B132
Interventions for prevention of CVC related infections in children with cancer: preliminary results of a Cochrane systematic review
Rebecca Roberts1, Ramandeep Arora2, Tim Eden3, Barry Pizer4
1University Hospital of South Manchester NHS Foundation Trust, Manchester, UK, 2University of Manchester, UK, 3Christie Hospital NHS Foundation Trust, Manchester, UK, 4Royal Liverpool Children's NHS Trust, Manchester, UK
Background
We looked at all interventions, other than anticoagulants and systemic antibiotics, to prevent CVC-related infections in children with cancer. We also examined effectiveness of each intervention across the following subgroups
(1) Implanted versus external catheters
(2) Haematological versus non-haematological malignancies
(3) Haematopoietic stem cell transplant(HSCT) patients versus non-HSCT patients
Method
A systematic search of electronic databases(MEDLINE, EMBASE, CINAHLplus & CENTRAL) was done. Reference lists of relevant articles and conference proceedings of ASCO, ASPHO, SIOP and MASCC(2004-2008) were also searched. Procedures for study selection, data extraction, and quality assessment were decided a priori. The primary outcome was catheter-related blood stream infection (CRBSI) and secondary outcomes were catheter associated infection (CAI), exit infection, tunnel infection, pocket infection and premature catheter removal for infection.
Results
The search strategy yielded 660 articles. We identified 6 RCTs for 5 different interventions: urokinase(2 studies), exit site dressing type(1 study), exit site dressing frequency(1 study), bacteriostatic saline flush(1 study) and blood reinfusion(1 study). Most studies had methodological limitations. No study reported on the primary outcome CRBSI. Among secondary outcomes, 2 studies (both had urokinase as intervention) reported on CAI with no significant difference in the control and study arms. 2 studies (1 had urokinase as intervention and the other had a longer exit site dressing frequency) reported premature catheter removal as an outcome and there were no significant differences in the control and study arms. No study reported on exit, tunnel or pocket infection. Because of limited number of studies and outcomes reported, meta-analysis was not done and subgroup analysis was not possible.
Conclusion
A lack of sufficient number of good quality RCTs reporting standard outcomes prevents us from making any definitive conclusions about the effectiveness of interventions for prevention of CVC related infection in children with cancer. Further research is clearly needed.
