Background: Catheter Related Blood Stream Infection (CRBSI ) and catheter colonization (CC) represent a significant problem encountered in hematopoietic stem cell transplantation (HSCT) setting adding to the morbidity, mortality and cost of healthcare in such patients. Unlike non-tunneled central venous catheters (CVCs), the higher complication rates and added cost associated with reinsertion of a tunneled CVC makes it important to devise effective strategies for catheter salvage.We report the incidence of CRBSI and CC of tunneled CVCs in our patients and use of antibiotic locks to salvage these catheters.

Materials and Methods:

All patients who underwent HSCT between September 2008 and December 2013 in our centre were included in this study. These patients underwent tunneled CVC (Hickman) insertion through internal jugular vein by Seldinger technique. Wherever possible, triple lumen Hickman catheters were used for patients undergoing allogeneic transplant, whereas double lumen catheters were used for those undergoing autologous transplant. Post insertion, surveillance blood cultures from all lumina of the catheter were taken. Catheter colonization (CC) was defined as growth of an organism from any lumen of the catheter without any evidence of fever as part of surveillance strategy. CRBSI was defined as growth of a recognized pathogen from catheter without growth from peripheral vein or growth from catheter with similar growth from peripheral vein with 2 hours lesser incubation for catheter in the presence of fever (>38o C), chills or hypotension. Antibiotic lock was used in all patients with CC according to the suceptibility patterns of the isolates, which included a third generation cephalosporin- β-lactamase inhibitor combination, aminoglycoside or carbapenem for gram negative bacteria; vancomycin, teicoplanin or linezolid for gram positive organism and amphotericin B or voriconazole for candida species. These patients did not receive systemic antibiotics. Systemic antibiotics were used in all cases of CRBSI. Catheters were removed in cases with persistent fever ( >72h), hemodynamic instability and repeated culture positivity. Antibiotic lock was continued for a total of 14 days with repeat cultures taken on 4th, 5th, 6th, 7th and 14th day of lock.

Results: Two hundred and twenty four patients (95 allogeneic and 129 autologous) underwent HSCT. The incidence of CC was 9.8 % (22/224) while CRBSI was seen in 10.7 % (24/224). Coagulase Negative Staphylococcus (CONS) and gram negative organisms (Pseudomonas aeruginosa-3, E.Coli-2, Acinetobacter baumanni-2, Klebsiella pneumonia-2 and Sphingobacterium thalpophilum-1) were seen in 10 each (45%) of CC. One case of Candida tropicalis causing CC was noted. Antibiotic lock was used in 21 of 22 CC cases. Catheter salvage was successful in 95% (20 / 21) cases. One case of CC progressed to CRBSI and required systemic antibiotics. Amongst CRBSI, gram negative bacteria accounted for 83% (20 / 24) of the cases ( Pseudomonas aeruginosa-6, Acinetobacter baumanni-4, E coli-4, Klebsiella pneumonia-3, Ralstonia picketti-3, Stenothrophomonas maltilophila-2, Sphingomonas paucimobilis-1 and Ochrobactrum anthropi-1). Forty-five percent of E. coli & Klebsiella pneumonia confirmed to be Extended Spectrum β-Lactamase (ESBL) producers. Vancomycin resistant enterococci (VRE) and Candida parapsilosis related CRBSI were seen in 1 patient each. Antibiotic lock was used for 15 of the 24 cases of CRBSI. Catheter salvage was successful in 86% (13 / 15) patients. Two patients had a second episode of CRBSI, after an interval of 50 and 10 days respectively from last documented negative culture. The organisms in both these cases were gram negative bacteria.

Conclusion: More than 80% of patients with catheter colonization and CRBSI could be salvaged with appropriate antibiotic locks in our setting. The use of antibiotic lock for tunneled CVCs is an effective strategy for catheter salvage in HSCT patients with CC and CRBSI.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution