ORLANDO—Patients with gram-negative peritonitis (GNP) undergoing early catheter removal have a higher rate of permanent transfer to hemodialysis but a lower death rate compared with GNP patients not having their catheter removed, according to study findings presented here at the 2008 Annual Dialysis Conference.


The study, led by Fred Finkelstein, MD, at the Renal Research Institute in New Haven, Conn., also showed that the risk of death is influenced by the type of gram-negative bacterial pathogen causing the infection. Co-investigator Laura Troidle, PA, presented study findings.

Continue Reading


The investigators prospectively studied episodes of GNP that developed from January 1, 1996 to August 31, 2007. The researchers collected data on causative organism, patient demographics, and outcome at three weeks and three months. Outcomes included death, transfer to hemodialysis, or return to peritoneal dialysis.


The researchers identified 139 GNP episodes. Pseudomonas was the most common causative pathogen (32 episodes), followed by Escherichia coli (26 episodes), Acinetobacter (19 episodes), Klebsiella (13 episodes), Enterobacter (10 episodes), Citrobacter (six episodes) and Serratia (eight episodes). Other gram-negative bacteria were isolated in 20 episodes. In nine episodes, only two gram-negative bacterial species were isolated.


Fifty-eight percent of patients who had their PD catheter removed were transferred permanently to hemodialysis compared with 12% of those who did not have their catheter removed, Troidle reported. Eight percent of those who had their catheter removed died compared with 24% who did not have their catheter removed.


Catheters generally were removed within five days of GNP onset if patients had continued cloudy fluid, per guidelines established by the International Society for Peritoneal Dialysis, Troidle noted. Patients who returned to peritoneal dialysis after catheter removal had their catheter reinserted immediately after the end of antibiotic treatment.

Additionally, GNP, especially with E. coli and Klebsiella, is associated with a high rate of technique failure and death within three months of GNP onset. At three months, 38% of patients with E.coli peritonitis and 23% of those with Klebsiella peritonitis had died.


“We feel that the unique biology of bacterial biofilm is responsible for the onset and outcome of gram-negative peritonitis,” Troidle said. “The rate of gram-negative peritonitis is low, fortunately. However, this creates a problem when looking at outcomes. As a result, Dr. Finkelstein and I have been interested in getting a multi-center database started to increase the [number of] episodes observed. It is necessary to track outcomes from a multicentered approach to better assess the outcome of GNP with regard to catheter removal and to determine if certain gram-negative organisms deserve more attention.”


Editor’s note: If you are interested in helping to establish a multicenter database for gram-negative peritonitis, please contact Laura Troidle, PA, at [email protected]