INTRODUCTION —
Catheter-related bloodstream infections (CRBSIs) occur in association with peripheral intravascular (IV) catheters and central venous catheters (CVCs) [1]. CVCs are increasingly used in the inpatient and outpatient setting to provide long-term venous access. Bloodstream infections (BSIs) associated with such catheters are often classified using the United States Centers for Disease Control and Prevention (CDC)'s National Healthcare Safety Network (NHSN) surveillance definition of central line-associated BSIs (CLABSIs) [2]. It is estimated that approximately 90 percent of CRBSIs in the United States occur with CVCs, though the role of peripheral IVs as causes of BSI is likely underestimated [1,3,4].
The epidemiology, risk factors, pathogenesis, and microbiology of IV catheter infections will be reviewed here. The diagnosis, treatment, and prevention of CVC-associated infections are discussed separately. (See "Intravascular non-hemodialysis catheter-related infection: Clinical manifestations and diagnosis" and "Intravascular non-hemodialysis catheter-related infection: Treatment" and "Routine care and maintenance of intravenous devices".)
EPIDEMIOLOGY —
In 2022, there were 23,389 central line-associated bloodstream infections (CLABSIs) reported by 3728 United States acute care hospitals to the United States Centers for Disease Control and Prevention (CDC)'s National Healthcare Safety Network (NHSN). The overall standardized infection ratio (SIR) is 0.84, representing a 16 percent decline compared with 2015 [5]. The SIR was higher in intensive care units (ICUs) than hospital wards (0.96 versus 0.78).
CLABSIs are an important cause of morbidity and mortality worldwide, although they may be decreasing in incidence in some areas, possibly as a result of widespread prevention efforts. As an example, the incidence of CLABSI associated with central lines among patients hospitalized in ICUs in the United States decreased from 3.64 to 1.65 infections per 1000 central line-days between 2001 and 2009 [6,7]. These decreases in the United States continued through 2015 and appear to be sustained [8,9], though the metric has switched from a rate to a SIR, and an increase in SIR was observed through the second year of the coronavirus disease 2019 (COVID-19) pandemic [10]. Higher CLABSI rates in ICUs were observed in 45 countries from 2015 to 2020, with a pooled mean of 4.55 infections per 1000 central line-days and with the highest rates observed in trauma ICUs and in low to middle income countries [11].
Although ICU patients are generally exposed to more medical devices and are more acutely ill than other hospitalized patients, CLABSIs remain common in hospital wards outside the ICU. As an example, among 23,389 CLABSIs reported to the CDC's NHSN, 9666 occurred among patients in adult ICUs, 12,449 in inpatient wards, and 1274 in neonatal ICUs in the United States in 2022 [12].
In a prospective analysis from the Surveillance and Control of Pathogens of Epidemiologic Importance database, which evaluated the characteristics of 24,179 hospital-onset bloodstream infections (BSIs) from 49 hospitals in the United States between 1995 and 2002, the following findings were noted [13]:
●The incidence of hospital-onset BSI was 60 cases per 10,000 hospital admissions. Approximately 15 percent of cases occurred in children.
●Approximately 51 percent of cases occurred in the ICU.
●Intravascular (IV) devices were the most common predisposing factor. A central venous catheter was in place in 72 percent, peripheral intravenous catheter in 35 percent, and an arterial catheter in 16 percent. (See 'Catheter factors' below.)
●The mean time from admission to BSI onset ranged from 12 to 26 days and depended on the isolated pathogen.
●The crude mortality rate was 27 percent.
Indwelling catheters are also a frequent source of infection in many populations who required long-term venous access, including hemodialysis and oncology patients, as well as those receiving total parenteral nutrition. In 2008 in the United States, an estimated 37,000 catheter-related BSIs (CRBSIs) occurred among patients receiving outpatient hemodialysis [6]. (See "Central venous catheters for acute and chronic hemodialysis access and their management" and "Tunneled hemodialysis catheter-related bloodstream infection (CRBSI): Management and prevention" and "Central venous catheters: Overview of complications and prevention in adults", section on 'Catheter-related infection'.)
BSIs associated with the widespread use of indwelling catheters have possibly contributed to the changing epidemiology of other infections. As an example, Staphylococcus aureus has emerged as a leading microbial cause of infective endocarditis in tertiary care centers, as illustrated by a study of a large international cohort of patients with both native and prosthetic valve endocarditis [14]. Infection with S. aureus versus other pathogens was independently associated with the presumed presence of an IV device. (See "Native valve endocarditis: Epidemiology, risk factors, and microbiology", section on 'Microbiology'.)
The distribution of pathogens associated with CRBSI is described below. (See 'Microbiology' below.)
RISK FACTORS
Host factors — Host factors commonly associated with hospital-onset bloodstream infections (BSIs) include the following [15,16]:
●Extremes of age
●Previous BSI
●Malnutrition
●Total parenteral nutrition administration
●Chronic illness
●Immune deficiency (especially neutropenia)
●Bone marrow transplantation
●Loss of skin integrity (as with burns)
Neutropenic patients are also at high risk for infections, including catheter-related BSIs (CRBSIs). Those with an absolute neutrophil count (bands plus polymorphonuclear leukocytes) <100 cells/microL appear to be at the greatest risk [17]. Hematologic malignancies have been associated more highly with central line-associated BSIs (CLABSIs) than solid tumors [18].
Among pediatric intensive care unit (ICU) patients, underlying gastrointestinal and oncologic diagnoses have also been associated with increased risk for CLABSI [19].
Patients hospitalized with burns are at particularly increased risk for hospital-onset BSIs as the necrotic tissue in the burn wounds increases the susceptibility to infection, thermal trauma decreases host resistance, and the body exhibits an inflammatory reaction to these processes [20].
Infection with COVID-19 has been shown in some studies to be associated with increased risk for CLABSI [21], including in national surveillance in the United States, which was initially associated with disruptions to routine infection prevention activities [22].
Catheter factors — All intravascular (IV) devices confer a risk of infection, although some (eg, nontunneled central venous catheters [CVCs]) carry greater risk than others (eg, peripheral venous catheters) [23].
For CVCs, the site of catheter placement affects the risk of infection, with the subclavian site being associated in some studies with less risk than other sites and therefore being the preferred insertion site in ICU patients [24,25]. Peripheral IV placement in the hand may be associated with lower risk of BSI compared with more proximal sites [26]. (See "Central venous catheters: Overview of complications and prevention in adults", section on 'Incidence by type of catheter' and "Central venous access: Device and site selection in adults", section on 'Factors influencing catheter selection' and "Central venous access: Device and site selection in adults", section on 'Anatomic location' and "Central venous access: Device and site selection in adults", section on 'Benefits/risk for specific sites'.)
Other than the type and location of the catheter, the most important extrinsic risk factors associated with the development of CRBSIs include:
●Duration of catheterization (although there is no indication for routine line changing based on number of catheter days)
●Type of catheter material
●Conditions of insertion
●Catheter-site care
●Skill of the catheter inserter
The importance of these factors varies with the site and type of IV catheter (table 1). In general, the risk of CLABSI is comparatively elevated in the following circumstances:
●Femoral or internal jugular placement compared with subclavian placement [27-30]
●Use for hyperalimentation or hemodialysis compared with other indications [15,31]
●Submaximal compared with maximal (mask, cap, sterile gloves, gown, large drape) barrier precautions during insertion [32,33]
●Nontunneled compared with tunneled insertion [34,35]
●Tunneled insertion compared with a totally implantable device [36-38]
●Bare compared with antibiotic-impregnated catheter [39]
●Multiple-lumen compared with single-lumen peripherally inserted central catheters (PICCs) [40]
Other factors that may increase the risk of infection include thrombosis of the catheter, repeated catheterization, increased manipulation of the catheter (including catheter repair), presence of infectious foci elsewhere, and disparities in the care of underrepresented groups [27,28,41-44]. A meta-analysis found that chemotherapy and immunosuppression were additional risk factors for CLABSI [45]. The use of antiseptic barrier caps for CVCs, CLABSI bundles, and blood diversion devices has been associated with reductions in CLABSI [46-48].
The incidence of BSI associated with peripheral venous catheters is low (0.04 per 1000 catheter days), but as over two billion peripheral IVs are estimated to be used worldwide every year, the absolute number of infections is high [49]. It has been estimated that one in three health care-associated S. aureus CRBSIs are due to peripheral intravenous catheters [3]. In a meta-analysis including more than 57,000 patients in 23 studies, PICCs were associated with a lower risk of BSI than CVCs (relative risk 0.62, 95% CI 0.4-0.94); however, among studies that reported BSIs per catheter day, PICC-related infection occurred as often as CVC infection (incidence rate ratio 0.91, 95% CI 0.46-1.79) [50]. Peripheral IVs with a ≤16 gauge diameter may be associated with a higher risk of BSI [51].
Detailed data describing modifiable catheter risk factors and methods of reducing the risk of infection are discussed elsewhere (see "Central venous access: Device and site selection in adults", section on 'Benefits/risk for specific sites'). Beyond host and catheter risk factors, institutional factors and safety culture may be associated with CLABSI [52].
SOURCES OF INFECTION —
Bloodstream infection (BSI) associated with central venous catheters (CVCs) can be attributable to four major sources: most often colonization from the skin or intraluminal or hub contamination, less often secondary seeding from a BSI, and, rarely, contamination of the infusate (figure 1) [53].
Skin colonization — The most common source of CVC-related infections is colonization of the intracutaneous and intravascular (IV) portions of the catheter by microorganisms from the patient's skin and occasionally the hands of health care workers (on insertion or as a result of manipulation) [54-56]. Furthermore, a number of studies have found a strong correlation between heavy skin colonization and both catheter colonization and subsequent catheter-related infection, especially with short-term IV devices [55,57-59]. Microorganisms gain access to the catheter wound and migrate along the catheter-subcutaneous tract into the fibrin sheath that surrounds IV catheters. Scanning electron micrographs reveal that both the external and internal surfaces of catheters can become colonized with microorganisms [60].
The deposition of biofilm on the external and internal surface of vascular catheters is thought to play an important role in the colonization process. The biofilm is produced by a combination of host factors (eg, fibrinogen and fibrin) and microbial products (eg, glycocalyx or "slime").
Thus, it is not surprising that common skin commensals, such as S. aureus and coagulase-negative staphylococci, are often isolated from colonized catheters and patients with primary BSIs. (See 'Microbiology' below.)
Intraluminal contamination — Intraluminal and/or hub contamination is an important source of BSI in patients with centrally inserted CVCs that are in place for more than two weeks or in patients with a surgically implanted device [61-64]. In one report in neonates, for example, catheter-related sepsis occurred with 26 percent of 113 catheters [64]. The infecting organism was grown from the catheter hub before or at the time of diagnosis of bacteremia in 54 percent of cases. A study from Spain found that a hub chamber containing iodinated alcohol significantly reduced both colonization and BSIs related to CVCs in intensive care unit (ICU) patients [65]. (See "Central venous access: Device and site selection in adults", section on 'Antimicrobial-impregnated catheters'.)
Hematogenous seeding — Hematogenous seeding of the device can occur during a BSI originating from another focus of infection, often from a gastrointestinal site; this is most likely to occur in critically ill patients or those with long-term catheters [27,57]. Recurrent bacteremia from the catheter may be difficult to distinguish from worsening of the primary focus of infection; removal of the catheter may help distinguish between these possibilities.
Infusate contamination — Administration of contaminated infusate or additives, such as a contaminated heparin flush, can result in a BSI. Organisms may contaminate infusate by several mechanisms: during manufacture, during solution preparation, via retrograde contamination from a contaminated catheter, or handling by health care workers. This is now a rare source of BSI and generally causes epidemic infections [66-70]. Use of diluted chlorhexidine via contaminated atomizers for cutaneous disinfection has similarly been associated with at least two outbreaks of BSIs and should be avoided [71].
Particular features can suggest the diagnosis. Infusate-related infections should be suspected when sepsis occurs in an otherwise low-risk patient receiving an intravenous solution or when there is a cluster of primary BSIs with an unusual organism, often an uncommon gram-negative bacillus. The infection is confirmed by the isolation of the same organism from both the infusate and percutaneous blood culture.
MICROBIOLOGY
General principles — Prior to the 1980s, gram-negative aerobes were the predominant organisms associated with hospital-onset bloodstream infections (BSIs). Since then, gram-positive aerobes (eg, coagulase-negative staphylococci, S. aureus, and Enterococcus) and Candida species have increased in relative importance [13,72].
As an example, the following distribution of pathogens was noted in the United States Centers for Disease Control and Prevention (CDC)'s National Healthcare Safety Network (NHSN) between 2011 and 2014 [73]:
●Coagulase-negative staphylococci – 16.4 percent
●S. aureus – 13.2 percent
●Enterococci – 15.2 percent
●Candida species – 13.3 percent
●Klebsiella species – 8.4 percent
●Escherichia coli – 5.4 percent
●Enterobacter species – 4.4 percent
●Pseudomonas species – 4 percent
Subsequent surveillance reports in the United States and Europe have documented similar microbiologic distributions, with coagulase-negative staphylococci, S. aureus, enterococci, and Candida spp accounting for the majority of catheter-related BSIs (CRBSIs) [74-76] and a trend toward more antimicrobial resistance among the pathogens [73]. The predominance of staphylococcal species, which are common constituents of the skin microbiome, reflects the observation that central venous catheter (CVC) infections are most commonly attributable to the patient's skin microbiome [77]. Besides staphylococci, enterococci, Candida spp, and Klebsiella spp are increasingly recognized as BSI pathogens in NHSN reports [78], with Pseudomonas spp, Acinetobacter spp, and more resistant organisms being important pathogens in intensive care unit (ICU)-onset infections [79].
A report from the CDC's NHSN from 2011 to 2017 demonstrated some changes in trends in the pathogens associated with 136,264 central line-associated BSIs (CLABSIs) [80]. Enterobacterales was the most commonly reported pathogen group in all sites except adult ICUs, for which the most commonly reported pathogen group was Candida spp. This analysis highlights the important role of Enterobacterales in CRBSIs.
Gram-negative bacilli may account from 16-31 percent of CLABSIs [13,73]. The most commonly isolated organisms include E. coli, Klebsiella pneumoniae, Pseudomonas species, Enterobacter species, Serratia species, and Acinetobacter species. Stenotrophomonas maltophilia bacteremia is a particular problem in cancer patients and is commonly CVC related [81]. A cluster of cases of catheter-related infections with gram-negative bacilli should prompt suspicion of an infusate contamination. (See 'Infusate contamination' above.)
BSIs caused by multidrug-resistant gram-negative bacilli have become a major concern because of increased rates of treatment failure and death. (See "Gram-negative bacillary bacteremia in adults", section on 'Antibiotic resistance'.)
Fungi, especially Candida species, account for up to 27 percent of CLABSIs [13,78,80]. Fungal infections are a particular concern in patients receiving a high concentration of glucose in intravenous hyperalimentation. Candida species also produce glycocalyx, which enhances their ability to colonize CVCs. Candida infections with CVCs occur more frequently in immunosuppressed patients and in patients who have received multiple antibacterial antibiotics [82]. Issues related to candidemia are discussed further separately. (See "Clinical manifestations and diagnosis of candidemia and invasive candidiasis in adults" and "Management of candidemia and invasive candidiasis in adults".)
BSI related to CVCs occasionally may be due to more than one microorganism [77]. CVC-related infection due to anaerobic bacteria is exceedingly rare.
Special populations — Certain patient populations have a different epidemiology of BSI with regard to microbial etiology:
●Among burn patients, Pseudomonas aeruginosa is the most frequently isolated gram-negative pathogen and, in one series, accounted for 16 percent of all cases of BSI [83]. In a prospective study of 45 bacteremic burn patients in India, Pseudomonas species were the most common isolates, responsible for 42 percent of cases [84].
●In patients with hematologic and nonhematologic malignancies, gram-negative pathogens predominate [85]. This association may be due to the translocation of gut bacteria in patients with altered mucosal barriers, and thus, the contribution of the catheter to the infection may be debatable [86].
●In hemodialysis patients, gram-positive organisms reflecting skin microbiome are responsible for most catheter-related infections. (See "Central venous catheters for acute and chronic hemodialysis access and their management" and "Tunneled hemodialysis catheter-related bloodstream infection (CRBSI): Management and prevention".)
●In infections associated with needleless access devices, hydrophilic gram-negative pathogens such as Pseudomonas species, Klebsiella species, Stenotrophomonas (Xanthomonas) species, Acinetobacter species, and Serratia marcescens appear to be the most frequently isolated pathogens [87-89]. It is thought that the devices are exposed to these pathogens during patient bathing or showering.
●In the ambulatory setting, children younger than three years or without recent hospitalization (within the week prior to presentation) may have an increased likelihood of polymicrobial CLABSIs [90].
●In patients, particularly very young infants, receiving intravenous lipids, Malassezia furfur, a lipophilic yeast, has been identified as a cause of BSIs [91].
●In patients receiving a high concentration of glucose in intravenous hyperalimentation, fungal infections, and Candida species in particular, are a particular concern.
Infusate contamination — BSIs related to infusion of contaminated fluids generally involve gram-negative bacilli, including Klebsiella, Citrobacter, or non-aeruginosa Pseudomonas species. Enterobacteriaceae, specifically, are usually associated with contaminated glucose-containing infusates. Fungi, especially Candida parapsilosis, are often associated with contaminated hypertonic parenteral nutrition infusate [53]. A cluster of cases of BSI involving the same gram-negative or fungal organism should prompt an investigation into the possibility of contaminated infusate. (See 'Infusate contamination' above.)
Local infection — Infections of the insertion or exit site, pocket, or tunnel are generally due to the same organisms listed above (see 'Microbiology' above). S. aureus and P. aeruginosa may be isolated in a greater proportion of these local infections than in CLABSIs.
INFORMATION FOR PATIENTS —
UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)
●Basics topic (see "Patient education: Central line infection (The Basics)")
SUMMARY AND RECOMMENDATIONS
●Epidemiology – Approximately half of hospital-onset bloodstream infections (BSIs) occur in intensive care units (ICUs), and the majority are associated with the presence of an intravascular (IV) device. Catheter-related BSIs (CRBSIs) are an important cause of morbidity and mortality worldwide, including resource-limited settings. (See 'Introduction' above and 'Epidemiology' above.)
●Host-related risk factors – Hospitalized patients with burns or neutropenia are at particularly increased risk for hospital-onset BSIs. Other host risk factors include extremes of age, immune deficiencies in general, chronic illness, and malnutrition. (See 'Host factors' above.)
●Catheter-related risk factors – All types of IV catheters pose significant but varying risk of infection. Duration of catheterization, catheter material, insertion conditions, and site care also impact the risk of catheter-associated infections (table 1). (See 'Catheter factors' above.)
●Sources of infection – BSI associated with central venous catheters can be attributed to four major sources: colonization from the skin, intraluminal or hub contamination, secondary seeding from a BSI, and contamination of the infusate (figure 1). Infections related to the administration of contaminated intravenous fluids are rare but should be suspected when bacteremia occurs in an otherwise low-risk patient or when there is a cluster of BSIs with unusual organisms. (See 'Sources of infection' above.)
●Microbiology – The skin microbiome is an important source of IV catheter infection. Coagulase-negative staphylococci and Staphylococcus aureus, common skin commensals, are common isolates from hospital-onset BSIs. Candida spp are the most frequent pathogen for central line-associated BSIs in adult ICUs. Gram-negative pathogens are increasing in frequency and predominate in patients with hematologic and nonhematologic malignancies. Infections with multidrug-resistant organisms are associated with high rates of treatment failure and death. (See 'Microbiology' above.)
ACKNOWLEDGMENT —
The UpToDate editorial staff acknowledges Jeffrey Band, MD, FACP, FIDSA, and Robert Gaynes, MD, who contributed to earlier versions of this topic review.