Prevalent central venous catheter (CVC) rates among hemodialysis (HD) patients in Canada remain high. In October 2006, we implemented a three-step multidisciplinary quality improvement project in our in-centre HD unit. The primary objective was to convert 50% of suitable patients to arteriovenous fistulas (AVFs) or arteriovenous grafts (AVGs). DESIGN, SETTING, PARTICIPANTS, AND MEASUREMENT: We undertook a case-crossover evaluation of the efficacy of a three-step conversion strategy. In step one, all medically suitable in-centre HD patients were assessed for arteriovenous (AV) access creation. In step two, patients were scheduled for preoperative vascular mapping and referred to the vascular surgeon. In step three, patients who refused conversion were asked to sign a waiver indicating that their decision to continue with a CVC was against medical advice.
At the start of the project in October 2006, there were a total of 284 patients on HD in our in-centre unit and 108 patients were catheter-dependent (38%). Of these, 53 patients were deemed suitable for conversion from a CVC to AVF or AVG; 26/53 (49%) patients agreed to conversion and 27/53 (51%) refused conversion. For the patients in the conversion group, 63% had been followed in chronic kidney disease (CKD) clinic and 37% initiated dialysis acutely; compared to 57% and 43% respectively in the refusal group. The difference was not statistically significant (p = 0.62 by Chi-square test), suggesting that there may be other factors affecting a patient's decision other than predialysis nephrology care. Of interest, 19/27 (70%) of patients who refused conversion signed the waiver and 8/27 (30%) refused to sign the waiver. None of the patients, when confronted with the waiver, agreed to conversion. Based on analysis of the main findings from our study, patients were most concerned about insertion of needles, pain and the appearance of their AV accesses. While 22 patients have successfully converted, resulting in a conversion rate of 41.5%, the percentage of catheter-dependent patients increased from 38% to 46% during the project period. Factors that likely contribute to the increase in point-prevalence CVC rates during the project period include a high rate of patient refusal, a high rate of patients deemed to be medically unsuitable, AV access failure during the project period, and most common was a failure to create AV access among incident HD patients who were followed in our centre through the late stages of chronic kidney disease (CKD). Successful conversion was defined as removal of CVC and use ofAVaccess for HD at the end of the study period (December, 2010).
Long-term CVC use in Canada and the unwillingness of medically suitable patients to convert to more optimal forms of vascular access are linked problems with potentially grave consequences. We need to develop a better understanding of the patients' perspective and possible psychological factors affecting patients' decisions if we are to have an impact on the high CVC use of Canadian prevalent HD patients.