Version May 2024
INTRODUCTION — In the United States, approximately 80 percent of patients initiate hemodialysis with a tunneled, cuffed dual-lumen catheter, and approximately 20 percent of all prevalent hemodialysis patients use such catheters [1,2]. Tunneled double-lumen catheters are generally used for short- and intermediate-term venous access in hemodialysis patients. These catheters are also used for permanent vascular access in some patients, particularly those with limited alternative options for vascular access [1]. (See "Central venous catheters for acute and chronic hemodialysis access and their management".)
Tunneled catheters are associated with a number of complications and, in particular, hemodialysis catheter-related bloodstream infection (CRBSI). The epidemiology, pathogenesis, clinical manifestations, and diagnosis of CRBSI are presented in this topic review. Treatment and prevention of CRBSI are presented elsewhere. (See "Tunneled hemodialysis catheter-related bloodstream infection (CRBSI): Management and prevention".)
Complications related to nontunneled catheters and noninfectious catheter-related complications, such as central vein stenosis and catheter thrombosis, are discussed separately:
●(See "Central vein obstruction associated with upper extremity hemodialysis access".)
●(See "Malfunction of chronic hemodialysis catheters".)
●(See "Central venous catheters: Overview of complications and prevention in adults".)
●(See "Central venous access in adults: General principles".)
EPIDEMIOLOGY AND RISK FACTORS — The majority of bloodstream infections in patients on hemodialysis are caused by infection of vascular access catheters. Most infections are due to skin flora colonizing the site around the site of the catheter placement [3]. The frequency of hemodialysis catheter-related bloodstream infections (CRBSIs) in several large case series has ranged between three and six episodes per 1000 catheter-days, which corresponds to an approximate incidence of one to two episodes of bacteremia per catheter-year [4]. In one study of 472 patients who had newly placed tunneled hemodialysis catheters, CRBSIs occurred in 35 percent by three months and 54 percent by six months [5].
CRBSIs among patients who have nontunneled catheters are two- to threefold more frequent compared with patients who have tunneled catheters, in part because of differences in catheter design [6,7]. As an example, the cuff on the outside of tunneled catheters forms a mechanical barrier that limits entry of skin flora along the outer wall of the catheter. By contrast, nontunneled catheters have no cuff, thereby allowing introduction of bacteria both through the catheter lumen and along the outer wall of the catheter.
The incidence of bacteremia is approximately 10-fold higher in patients with tunneled catheters compared with patients who have either arteriovenous (AV) fistulas or grafts [8-10]. In addition, catheter-dependent hemodialysis patients have a two- to threefold higher risk of infection-related hospitalization and infection-related death compared with patients undergoing hemodialysis via an AV fistula or graft [11,12].
The clinical impact of catheter use in hemodialysis patients has been quantified in several studies [8,13-15]. As an example, in the Hemodialysis (HEMO) study, 7.6 percent of all patients had catheters used for vascular access, yet this group comprised 32 percent of all study patients hospitalized with an access-related infection [15]. In another prospective study of 2666 Scottish patients on kidney replacement therapy, use of a tunneled catheter for hemodialysis access was associated with a sevenfold higher odds of death compared with patients using AV fistulas or grafts [16]. Yet, despite these risks and international guidelines favoring early placement of AV fistulas to minimize need for tunneled catheters, the use of tunneled catheters has steadily increased in many countries [17].
Hemodialysis CRBSIs can lead to metastatic complications such as osteomyelitis, endocarditis, septic arthritis, or epidural abscess. Metastatic infections have been observed in approximately 5 to 10 percent of catheter-dependent hemodialysis patients [18]. The increased frequency of catheter use has led to more frequent metastatic infections [19].
The risk of metastatic infection varies with the type of infecting pathogen. As an example, serious metastatic infections are more frequently due to Staphylococcus aureus, which accounts for 10 to 40 percent of such infections [20,21]. In one large series, a metastatic infection occurred in 10 percent of patients with CRBSI caused by S. aureus versus 4 percent of patients who had Staphylococcus epidermidis or Gram-negative infection [18]. It is important to note that metastatic infections may not be immediately apparent, as they can first manifest clinically weeks or months after the initial bacteremic event.
The most important risk factor for tunneled CRBSIs is prolonged use of the catheter [5,22]. Other risk factors include a history of previous CRBSIs, recent surgery, diabetes mellitus, iron overload, immunosuppression, and hypoalbuminemia [4,23-25].
MICROBIOLOGY AND PATHOGENESIS — Gram-positive organisms are responsible for most hemodialysis catheter-related infections. Coagulase-negative staphylococci and S. aureus together account for 40 to 80 percent of cases in most studies [4,20,26-30]. Gram-negative organisms account for 20 to 40 percent, and polymicrobial infections are implicated in 10 to 20 percent of all episodes of catheter-related bloodstream infections (CRBSIs) [4].
S. aureus infection is commonly associated with significant morbidity and mortality. (See "Clinical manifestations of Staphylococcus aureus infection in adults" and "Epidemiology of Staphylococcus aureus bacteremia in adults".)
As examples:
●In a retrospective cohort study of 22,130 hospitalizations of hemodialysis patients with septicemia, the overall death rate from S. aureus bacteremia after 12 weeks of follow-up was 34 percent [13]. The death rate was 20 percent higher than the death rate from bacteremia due to all other organisms.
●In another study, 21 percent of 11,572 admissions of hemodialysis patients for S. aureus bacteremia had one or more complications [14]. The average length of first hospitalization was 13 days, and 12 percent of patients needed a readmission to the hospital within 12 weeks for treatment of a relapsed infection.
However, the death rate actually attributable to S. aureus bacteremia is lower. As an example, in a series of 84 patients with S. aureus bacteremia, only 2 percent had death directly caused by the CRBSI or its metastatic complications [18].
Methicillin-resistant S. aureus (MRSA) infection has also become an important pathogen in this population [31]. (See "Methicillin-resistant Staphylococcus aureus (MRSA) in adults: Epidemiology", section on 'Presence of an indwelling hemodialysis catheter'.)
Non-staphylococcal hemodialysis catheter-related bacteremias are predominantly due to enterococci and gram-negative rods [4,20,26-29,32,33]. Human immunodeficiency virus (HIV)-positive hemodialysis patients are more likely to develop polymicrobial infections or infections due to gram-negative or fungal pathogens [34,35].
The pathogenesis of hemodialysis CRBSI is similar to that of other catheters. Hemodialysis CRBSIs can arise from one of two sources [4] (see "Intravascular catheter-related infection: Epidemiology, pathogenesis, and microbiology", section on 'Sources of infection'):
●Migration from the skin along the outside of the catheter into the bloodstream. The Dacron cuff in tunneled catheters limits this migration of bacteria by inciting a cascading inflammatory response ending in fibrotic tissue that serves as a mechanical barrier against the skin flora.
●Direct inoculation from biofilm-containing pathogenic micro-organisms that may form on the inner surface of the catheter.
CLINICAL MANIFESTATIONS — Although nonspecific, fever and/or chills are the most useful clinical clues to the presence of a catheter-related bloodstream infection (CRBSI). In three prospective clinical studies, the presence of fever or chills in catheter-dependent hemodialysis patients was associated with positive blood cultures in a substantial proportion of patients [33,36,37].
Most cases of hemodialysis CRBSIs occur in the absence of an exit-site infection (purulence at the insertion site). However, the presence of an exit-site infection strongly predicts the concurrent presence of a bloodstream infection. In a series of 1436 episodes of tunneled hemodialysis catheter-related bacteremias, a concurrent purulent exit-site infection was observed in only 4.6 percent of cases, including 9.6 percent of those with a S. epidermidis infection, 6.1 percent of those with an S. aureus infection, and <1 percent of those with a gram-negative infection [38].
Less common clinical manifestations of a CRBSI may include hemodynamic instability, altered mental status, catheter dysfunction, and other signs of sepsis, such as hypothermia, acidosis, and hypotension.
Complications from CRBSIs, such as suppurative thrombophlebitis, endocarditis, septic arthritis, osteomyelitis, or abscess, may be the first manifestations of infection before a catheter infection becomes apparent. The clinical manifestations of these complications are discussed at length elsewhere:
●(See "Septic arthritis in adults", section on 'Clinical manifestations' and "Cellulitis and skin abscess: Epidemiology, microbiology, clinical manifestations, and diagnosis", section on 'Skin abscess'.)
●(See "Septic arthritis in adults", section on 'Clinical manifestations' and "Cellulitis and skin abscess: Epidemiology, microbiology, clinical manifestations, and diagnosis", section on 'Skin abscess'.)
DIAGNOSIS
When to suspect a catheter-related bacteremia — Catheter-related blood stream infection (CRBSI) should be suspected in any patient with a hemodialysis catheter and signs and/or symptoms of a systemic infection, particularly when there is no clinical evidence for an alternate source. (See 'Clinical manifestations' above.)
The diagnostic approach begins with a basic history and physical examination. Indications for additional laboratory and radiologic evaluation depend upon the findings of this evaluation. A CRBSI is confirmed in approximately 85 percent of hemodialysis patients who have a catheter and a suspected infection. Thus, an alternate source of infection is identified in only about 15 percent of such cases [18,39].
How to confirm diagnosis — Among patients with a clinical suspicion of CRBSI, a diagnosis is confirmed by two blood cultures obtained prior to administration of antimicrobial therapy and after exclusion of alternate sources of infection. Blood cultures may be obtained in various combinations from sites including the hemodialysis catheter, hemodialysis circuit, or peripheral veins. Ideally, at least one of the cultures should be obtained from a peripheral vein. However, peripheral vein cultures are infrequently obtained in hemodialysis patients, and hemodialysis circuit cultures are typically drawn instead.
A minimum of 10 mL of blood should be obtained from each site. We do not routinely rely on catheter tip culture for diagnosis of CRBSI, as removal of a tunneled catheter at the first suspicion of infection is only indicated in certain circumstances. Patients who have an exit-site infection should have any visible purulent drainage sent for culture.
Tests for other potential sources of infection, such as a chest radiograph or urine culture, should be individualized based upon the initial evaluation of the patient (ie, a complete review of systems for symptoms of infection). Among patients with localizing features, appropriate investigations to evaluate the alternate source should be undertaken (eg, chest radiograph for a patient with productive cough). However, among patients with nonspecific systemic symptoms and signs of an infection (eg, fever, chills, malaise) without other localizing features, limiting the evaluation to CRBSI by obtaining blood cultures is usually sufficient.
Our choice of the site(s) for blood culture samples depends upon timing of onset of symptoms relative to the timing of the dialysis procedure (see "Detection of bacteremia: Blood cultures and other diagnostic tests", section on 'Collecting blood cultures'):
●If a patient develops symptoms shortly before their next hemodialysis session (eg, within a few hours), then we obtain one blood culture sample from the hemodialysis catheter and another one from either a peripheral vein or the hemodialysis circuit bloodline while the patient is on dialysis. Generally, patients present to their next dialysis session with a history of symptom onset in the interdialytic interval.
●If a patient develops symptoms or signs of a bloodstream infection during a hemodialysis treatment, then we obtain two samples from the hemodialysis circuit bloodline [40] or one sample from the hemodialysis circuit bloodline and another one from a peripheral vein.
●If a patient develops symptoms or signs of a bloodstream infection during the 48- to 72-hour interdialytic interval, but not shortly before their next hemodialysis session (eg, within a few hours), then we obtain one sample from the hemodialysis catheter and another one from a peripheral vein. An alternative is to obtain two samples from peripheral veins. Among patients who are unable to have a peripheral vein sampled, we obtain two samples from the hemodialysis catheter 10 to 15 minutes apart.
Several factors influence the preference by many clinicians for hemodialysis circuit cultures rather than peripheral vein cultures [4,41-43]:
●Absence of suitable peripheral veins
●The desire to avoid venipuncture, even if suitable peripheral veins are present, to preserve them for a potential future fistula or graft
●Lack of sufficient dialysis nurse proficiency in obtaining peripheral vein cultures, and the absence of phlebotomy teams in the outpatient dialysis setting
●If symptoms occur during a hemodialysis session, it is unlikely that there is a meaningful difference between samples drawn from peripheral veins and those drawn from the hemodialysis tubing while systemic blood is rapidly circulating through the hemodialysis circuit
Thus, among patients who develop symptoms shortly before or during a hemodialysis session, circuit bloodline cultures are a reasonable alternative to peripheral blood cultures in many dialysis patients [43].
The definitive diagnosis of CRBSI requires concurrent blood cultures positive for the same organism from two sites, in the absence of an alternate source of infection. Additional criteria that may be relevant, depending upon the type of organism involved, are presented in detail elsewhere. (See "Intravascular non-hemodialysis catheter-related infection: Clinical manifestations and diagnosis" and "Intravascular non-hemodialysis catheter-related infection: Clinical manifestations and diagnosis", section on 'Interpreting blood culture results'.)
PREVENTION — The prevention of hemodialysis CRBSI is similar to prevention of other infection due to other central vascular lines. (See "Central venous catheters: Overview of complications and prevention in adults".)
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Dialysis".)
SUMMARY AND RECOMMENDATIONS
●Epidemiology and risk factors – The majority of bloodstream infections in patients on hemodialysis are caused by infection of vascular access catheters. The incidence of hemodialysis catheter-related bloodstream infections (CRBSIs) is approximately one to two episodes per catheter-year. Metastatic infections have been observed in approximately 5 to 10 percent of catheter-dependent hemodialysis patients, most frequently in association with Staphylococcus aureus infections. (See 'Epidemiology and Risk Factors' above.)
●Microbiology and pathogenesis – Gram-positive organisms are responsible for most CRBSIs. Coagulase-negative staphylococci and S. aureus together account for 40 to 80 percent, gram-negative organisms for 20 to 40 percent, and polymicrobial infections for 10 to 20 percent of all episodes of CRBSIs. The pathogenesis of hemodialysis CRBSI is similar to that of other catheters. It results from migration of bacteria along the outside of the catheter into the bloodstream or by direct inoculation from biofilm-containing pathogenic micro-organisms, which may form on the inner surface of the catheter. (See 'Microbiology and Pathogenesis' above.)
●Clinical manifestations – Fever and chills are the most useful clinical clues to the presence of CRBSI. Most cases of hemodialysis CRBSIs occur in the absence of a concomitant exit-site infection. Less common clinical manifestations of a CRBSI may include hemodynamic instability, altered mental status, catheter dysfunction, and other signs of sepsis, such as hypothermia, acidosis, and hypotension. (See 'Clinical manifestations' above.)
●Diagnosis – CRBSI should be suspected in any patient with a hemodialysis catheter and signs and/or symptoms of a systemic infection, particularly when there is no clinical evidence for an alternate source. The diagnosis of CRBSI is confirmed by two blood cultures obtained prior to administration of antimicrobial therapy and after exclusion of alternate sources of infection. The site of blood culture sampling depends upon timing of onset of symptoms relative to the timing of the dialysis procedure (see 'Diagnosis' above):
•If a patient develops symptoms shortly before their next hemodialysis session (eg, within a few hours), then we obtain one blood culture sample from the hemodialysis catheter and another one from either a peripheral vein or the hemodialysis circuit bloodline while the patient is on dialysis.
•If a patient develops symptoms or signs of a bloodstream infection during a hemodialysis treatment, then we obtain two samples from the hemodialysis circuit bloodline or one sample from the hemodialysis circuit bloodline and another one from a peripheral vein.
•If a patient develops symptoms or signs of a bloodstream infection during the 48- to 72-hour interdialytic interval, but not shortly before their next hemodialysis session (eg, within a few hours), then we obtain one sample from the hemodialysis catheter and another one from a peripheral vein. An alternative is to obtain two samples from peripheral veins. Among patients who are unable to have a peripheral vein sampled, we obtain two samples from the hemodialysis catheter 10 to 15 minutes apart.
آیا می خواهید مدیلیب را به صفحه اصلی خود اضافه کنید؟
