The use of tunneled dialysis catheters to deliver hemodialysis treatment may be associated with major problems. For this reason their use should be minimized as much as possible. Infection is the most serious of these problems. This complication causes very significant morbidity and mortality and has emerged as the primary barrier to long-term catheter use. Bacteremia is the most serious type of infection associated with catheter use. It can result in metastatic infection and even lead to death of the patient. Prophylaxis is important to decrease the risk of infection. The use of an antibiotic ointment at the exit site until it has healed and the long-term use of a dressing to cover the exit site are effective in decreasing the incidence of exit-site infection. With optimal catheter-use management, it should be possible to reduce the incidence of catheter-related bacteremia (CRB) to a level in the range of 1/1000 catheter days. Antibiotic and antimicrobial locking solutions show promise and may, if verified in appropriate clinical studies, prove to be important adjuncts to the management of catheter-dependent patients. Aspirin has been shown to have anti-staphylococcal activity and warrants further clinical evaluation. The diagnosis of CRB is based upon positive blood cultures in association with typical clinical features. If a simple routine blood culture is positive, along with a high clinical probability based upon the patient's signs and symptoms, the sensitivity and specificity of the diagnosis is greater than 75%. CRB is in reality a biofilm infection and must be treated as such. Treatment needs to focus on appropriate systemic antibiotics which should be continued for a minimum of 3 weeks and catheter management to remove the biofilm. Catheter exchange has been shown to be effective and should be performed based upon the clinical presentation of the patient. While treatment with a combination of systemic antibiotics and antibiotic locking solution may be effective for gram-negative infections, this approach does not appear to be a good choice for Staphylococcus aureus CRB.