Prevention of microbial colonisation of tunnelled haemodialysis catheters with Cathasept lock solution: multi-centre, prospective, randomised clinical trial of efficacy and safety

Infection is the second common cause of death in the dialysis population. Catheter- related blood stream infections (CRBSI) are associated with increased morbidity and mortality in haemodialysis (HD) patients. The use of tunnelled haemodialysis catheters (t-HDC) seems to be rising in the western world. The management of CRBSI pose a great challenge to nephrologists due to the lack of a gold standard diagnostic method or a widely accepted treatment strategy. Endoluminal colonisation is regarded as the main mechanism for blood stream infection in patients with t-HDC, and may result in sub- clinical inflammation with adverse effects on anaemia control in HD patients. The use of anti-microbial lock solutions has been shown to reduce the rate of CRBSI in several studies. This thesis reports data derived from the Cathasept trial which compared the effectiveness of a novel anti-microbial lock solution made of Tetra-sodium Ethylene Diamine Tetra-acetic Acid 4% (EDTA) with the standard heparin lock solution in the prevention of microbial colonisation of t-HDC and its impact on CRBSI incidence rate and on inflammation and anaemia parameters in 117 haemodialysis patients with confirmed uncolonised t-HDC. The study showed a significant reduction in catheter colonisation rate in patients allocated to the Cathasept arm but a non-significant reduction in CRBSI rate. Cathasept was associated with more thrombotic complications and with reduced catheter patency rate compared to heparin. Despite lower catheter colonisation and CRBSI episodes, Cathasept was associated with higher C-reactive protein (CRP) levels. There was no difference between both groups with regards to haemoglobin or ferritin levels with similar iron and darbepoetin requirements. This thesis also reports data from fiver laboratory-based experiments. The impact of prolonged incubation time (48hrs vs 24 hrs as per clinical study protocol) and using 3 blood agar culture medium as opposed to C.L.E.D agar which was used in the clinical study to perform through-catheter quantitative blood culture (TCQBC) was assessed in a separate experiment with no improvement in the culture yield. In another experiment I was unable to prove that quantitative culture of heparin lock solution obtained from the catheter lumen before obtaining the blood sample yields more positive culture results. Endoluminal brushing was carried out on 23 explanted t-HDC. This technique was able to confirm colonisation of four catheters with previous negative TCQBC but, on the other hand, it was negative in three catheters which had yielded positive TCQBC. Forty four segments (0.5 cm each) from 11 ex-vivo t-HDC were examined by SEM prior to endoluminal brushing and quantitative culture to determine their colonisation status. Catheter colonisation rate detected by SEM was 36% which is in contrast to the findings of a previous study which reported universal colonisation of long-term HD catheters. I attempted to study the viability of biofilms generated on PVC tubes using confocal laser scanning microscopy (CLSM) but this experiment was unsuccessful due to auto- florescence of the catheter surface preventing the visualisation of bacteria when SYT09 stain was used, and the failure of Propidium iodide stain to penetrate biofilms which made it impossible to distinguish live and dead cells in the biofilm. 4