The most common treatment for patients with chronic kidney failure is hemodialysis, requiring reliable access to the bloodstream. There are three types of vascular access: arteriovenous fistulas (AVFs), arteriovenous grafts (AVGs), and central venous catheters (CVCs). AVFs are strongly recommended in international guidelines; guidelines also discourage the use of CVCs for long-term vascular access.
However, the use of CVCs worldwide is as high as nearly 70% and ranges from 15% in the United States to 45% in Canada among prevalent hemodialysis patients. Canada uses more CVCs than other countries, yet there are few available data on the risk profile associated with CVC use among Canadian patients.
Patients may opt for a CVC due to ineligibility for an AVF or AVG determined by their medical team or because they resist having a fistula or graft created. Elderly patients are more likely to be considered ineligible due to a high burden of comorbid conditions that add to surgical risk, a shortened life expectancy, and/or poor vasculature increasing the risk of primary failure. A previous study found that the rate of bloodstream infections from CVCs was one-third the normal rate in elderly patients; however, age is not a consistent predictor of lower risk of infection.
The current guidelines do not account for age in the recommendations for access type, other than to note that CVCs are acceptable if the duration of dialysis therapy is expected to be less than 1 year. Krishna Poinen, MD, and colleagues conducted an observational retrospective cohort study to examine the risk for complications in patients who use CVCs for vascular access and to assess the impact of age on the risk for complications. Results of the study were reported in the American Journal of Kidney Diseases [2019;73(4):467-475].
The researchers identified patients ≥18 years of age who initiated hemodialysis therapy and used a tunneled CVC between January 1, 2004, and May 31, 2012. Patients were included if they initially used a nontunneled CVC (eg, inpatients), but were switched to a tunneled CVC within 120 days. Exclusion criteria were life expectancy <1 year due to metastatic cancer or other terminal illness, transitioning to peritoneal dialysis within 6 months of initiation of dialysis therapy, or never receiving outpatient dialysis because the CVC use would be of a limited duration. Other exclusion criteria were creation of an AVF or AVG after initiating hemodialysis therapy with a tunneled CVC.
Follow-up continued from the index date (first use of tunneled CVC) for 2 years or until end of the study period (August 31, 2012) or the first of recovery of kidney function, kidney transplantation, transition to peritoneal dialysis therapy, withdrawal of care, transfer to another program, loss to follow-up, or death. Maximum follow-up was 2 years.
Exposures of interest were age, sex, body size, initiating dialysis therapy in the hospital, and comorbid conditions. The exposures were described for four groups: age <60 years, 60 to 69 years, 70 to 79 years, and ≥80 years at the time of first use of the tunneled CVC.
The outcomes of interest were procedures, hospitalizations, and deaths related to the CVC. The three complications were further categorized into five primary causes: bacteremia, other infections, central venous stenosis or thrombosis, restricted blood flow, or other CVC-related complication.
At the five Canadian centers participating in the study, 3145 patients initiated long-term dialysis therapy. Of those, 1041 used CVCs and met study inclusion criteria. Patients who only used CVCs were on average 2 years older and, excluding diabetes, had more comorbid conditions compared with patients who had had a fistula created. Patient age was evenly distributed in the CVC-only group. Younger patients were less likely to have coronary artery disease, congestive heart failure, cerebral vascular disease, peripheral vascular disease, and cancer. Older patients were less likely to have diabetes and lower body mass index (BMI).
Median follow-up from first tunneled CVC insertion was 241 days; length of follow-up was similar across age groups. Older patients were more likely to exit the study due to death, and transplantation was more common in younger patients.
The cumulative risk for any complications related to CVC use was 30% at 1 year and 38% at 2 years. A total of 318 patients experienced a complication related to CVC use. Of those, 76 (23%) had two complications and 114 (36%) had three or more complications. The majority of complications resulted in procedures (n=720). Hospitalizations related to CVC use were infrequent (n=77); death related to CVC use was rare (n=6).
The 1-year risk for bacteremia was 9%; the 2-year risk was 11%. There were 186 cases of CVC-related bacteremia. Of those, 56 patients were hospitalized and 184 had the CVC removed. The two cases without catheter removal were admitted with a diagnosis of CVC-related bacteremia. Bacteremia accounted for 72% of hospitalizations related to CVC use, and it contributed to death in four of the 1041 patients.
In recurrent-events model analysis, there was an association between older age and decreased risk for complications. Compared with patients younger than 60 years, the hazard ratio (HR) for CVC complications for patients 70 to 79 years of age was 0.67 (95% confidence interval [CI], 0.52-0.85; P=.001); for patients 80 years of age and older, the HR was 0.69 (95% CI, 0.52-0.92; P=.01). Increasing BMI was associated with higher risk of CVC-related complications (HR for each 5-unit increase, 1.07; 95% CI, 1.01-1.12); inpatient hemodialysis initiation was also associated with higher CVC-complication risk (HR, 1.31; 95% CI, 95% CI, 1.05-1.62). There were no significant associations between higher CVC-related complication risk and sex, coronary artery disease, diabetes, or cancer.
Limited generalizability of the findings and lack of comparison of CVC use with other hemodialysis vascular access types were cited as limitations to the study.
In summary, the researchers said, “Approximately one-third of patients in Canada who used tunneled hemodialysis CVCs during 1 to 2 years experienced a CVC-related complication. Some complications such as bacteremia were serious but rarely directly led to patient death. Older patients were at lower risk for complications. This information can be used to better communicate the risk of CVC use in selected patients. Future studies comparing risks for access-related complications in patients choosing between AVFs and CVCs are needed, particularly for older patients.”
- Canadian researchers conducted a observational retrospective cohort study to examine the risks associated with central venous catheter (CVC) use, particularly in older patients.
- At 1-year of CVC use, the risk for any CVC-related complications was 30%; at 3-years, the risk was 38%.
- Compared with patients <60 years of age, patients ages 70 to 79 and those >80 years of age had lower rates of CVC-related complications (hazard ratio, 0.67 and 0.69, respectively).