Intravenous (i.v.) catheters are an almost mandatory part of hospital-based medical practice, providing necessary vascular access for i.v. fluids, drugs, blood products and total parenteral nutrition. At least 5 million peripheral vein i.v. catheters are sold every year in Australia,1, and it is rare for a patient to be admitted to hospital in Australia or New Zealand and not receive at least one i.v. catheter during their stay. However, their use is not without risk, as illustrated by the study reported in this issue of the Internal Medicine Journal of cannula-related Staphylococcus aureus bacteraemia (CASAB) in six New Zealand hospitals2. Apart from bacteraemia, other commonly reported i.v. line-related adverse events include local infections, septic thrombophlebitis and bacteraemic seeding of other sites including endocarditis3. Staphylococci are the commonest causative pathogens of i.v. line-related sepsis, but other organisms have been implicated including other Gram-positive bacteria, Gram-negative bacteria and fungi1. Six types of i.v. catheter are in common usage in Australasian hospitals. These include peripheral i.v. catheters, central venous catheters (CVC), peripherally inserted central venous catheters (PICC lines, which are inserted in the upper arm veins under image intensifier control or into the antecubital fossa), midline catheters (which are inserted at the same site as a PICC line, but are much shorter and hence do not enter the superior vena cava), tunnelled CVC, such as Hickman catheters and finally, ports that are totally implanted beneath the skin and accessed with a needle (the latter two types are primarily used in chemotherapy patients). In many hospitals nowadays, PICC lines are much more commonly inserted than jugular or subclavian CVC because of the lower complication rate at the time of insertion and also because they are the most convenient form of i.v. access in patients on home i.v. programmes. The study by Thomas and Morris in this issue primarily includes cases of bacteraemia associated with peripheral i.v. catheters and CVC (jugular and subclavian), because PICC lines were rarely used in the six New Zealand hospitals at the time of the study (1996–1999)2. In the United States, hospital-acquired bacteraemias occur at a rate of 1.3–14.5 per 1000 hospital admissions and are believed to directly result in 62 500 additional deaths per year4. Infected i.v. catheters are the commonest cause of hospital-acquired bacteraemias. In an Australian study, 42% (54/128) of hospital-acquired bacteraemias were a direct result of an infected intravascular device5. In 1994, it was estimated that at least 3000 cases of bacteraemia associated with i.v. catheters occurred in Australia1. In the same study, there was a 24% mortality in the 111 i.v. catheter-related bacteraemia patients who underwent prolonged follow up, although only half of these deaths were directly attributable to sepsis. In addition, it is important to note that infection rates differ between the various types of i.v. catheters. In 10 public hospitals enrolled in the New South Wales Hospital Infection Standardized Surveillance (HISS) programme, the rate of i.v. device-related bloodstream infections for all lines was 2.1 per 1000 line days, with the rate for central lines being 4.7 per 1000 line days, a rate much higher than that of peripheral i.v. lines6. The commonest and most important pathogen causing i.v. catheter-related bacteraemia is S. aureus, accounting for 149/240 (62%) episodes of peripheral i.v. catheter-associated bacteraemia and 145/531 (27%) of episodes of CVC-associated bacteraemia reported from 15 predominantly tertiary referral Australian hospitals in 19941. In over 90% of all episodes of peripheral i.v. catheter-associated sepsis in this study the catheter had been in situ for 3 days or more, and in all cases there was clinical evidence of thrombophlebitis. In contrast, in most episodes of CVC-related sepsis in the study there was no appreciable evidence of local infection. What is the economic cost of CASAB? In 21 US cases of CASAB, the median increase in length of stay for methicillin-susceptible S. aureus bacteraemia was 4 days at an extra cost of US$9661, compared to 12 extra days at an additional cost of US$27 083 for methicillin-resistant S. aureus (MRSA) bacteraemia7. In this study, there were no cases of endocarditis or other metastatic foci of infection complicating the cases of bacteraemia; if these had occurred there would have been a marked increase in costs and length of hospital stay. The pathogenesis of vascular catheter infections has recently been extensively reviewed8. To summarize, it has been conclusively shown that vascular catheters uniformly develop biofilms within a short time after insertion9. These biofilms are readily observed on electron microscopy and contain between zero and 10 million organisms. Studies in rabbits and humans have shown a direct correlation between the number of viable organisms, the likelihood of a positive blood culture and the severity of infection10. How then does the catheter become colonized with organisms? Three organisms, S. aureus, Staphylococcus epidermidis and Candida albicans are disproportionately common causes of catheter-associated intravascular infections. S. epidermidis produces slime that aids its survival on foreign bodies, and S. aureus produces a significant number of virulence factors, including proteins that interact with the coagulation system and with the extra-cellular matrix, which make it uniquely suited to survive within blood vessels. C. albicans can grow well in hyperalimentation fluid (the fluid used in total parenteral nutrition) and also adheres well to extracellular matrix proteins such as fibrinogen and fibronectin8. It is also known that foreign bodies such as catheters are associated with neutrophil dysfunction, including less effective phagocytosis and less effective killing around the catheter11. It is critical to the effective prevention of i.v. catheter-related infections that extrinsic factors that predispose to infection are considered because many of these factors can easily be altered. The density of the skin flora at the insertion site is a major risk factor for infection; hence, for peripheral i.v. cannulae, lower limb insertion sites are associated with a higher risk of infection than upper limb sites12. For CVC, femoral sites in adults have a much higher infection rate than jugular or subclavian CVC and are much more likely to predispose to deep venous thrombosis13. In addition, Teflon or polyurethane catheters are associated with fewer infectious complications than catheters made of polyvinyl chloride or polyethylene14. Many i.v. catheter-related infections could be prevented by scrupulous attention to good hand hygiene during insertion and catheter management. For short peripheral cannulae, the Centers for Disease Control and Prevention Guidelines recommend either a waterless alcohol-based handrub or washing with an antibacterial soap and water with adequate rinsing3. In addition, peripheral cannulae should be inserted with an aseptic technique using a new pair of gloves. Disposable non-sterile gloves are adequate provided a ‘no-touch’ technique is used during insertion. Gloves are also routinely recommended during insertion of cannulae to protect against exposure to blood-borne pathogens3. Compared with peripheral cannulae, CVC carry a much greater risk of infection and, hence, using maximal standard barrier precautions (mask, sterile gown, large sterile drapes and sterile gloves) during the insertion substantially reduces the risk of bacteraemia compared with standard precautions with sterile gloves and small drapes15. What skin antiseptic should be used prior to the insertion of i.v. catheters? In one large study, preparation of central venous and arterial sites with 2% aqueous chlorhexidine skin preparation lowered bacteraemia rates compared with site preparation with 10% povidone iodine or 70% alcohol16. This result is believed to be due to the longer residual activity of chlorhexidine on skin compared to iodine or alcohol. However, 2% aqueous chlorhexidine is not commercially available in Australia so many Australian hospitals use preparations containing a lower concentration of chlorhexidine (0.5–1%) combined with 70% alcohol prior to insertion of i.v. catheters. What dressings should be used on catheter sites? A meta-analysis of studies comparing the risk of catheter-related bacteraemia for groups using transparent semipermeable polyurethane dressings versus groups using gauze dressings showed that the rates of catheter-related bacteraemia did not differ between the two groups17. However, two practical advantages of transparent dressings are that the catheter site is constantly on view and that patients can shower and bathe without saturating the dressing, saving on nursing time. There has been a recent trend to coat CVC externally and now also internally with antiseptics and/or antibiotics to try to reduce the rates of catheter-related bacteraemia. Two meta-analyses have demonstrated that CVC coated with chlorhexidine/silver sulfadiazine did substantially reduce the risk of bacteraemia compared to standard non-coated CVC18,19. Disadvantages of these impregnated catheters include increased cost and anaphylaxis which, although rare, has been reported20. In addition, whether organisms resistant to chlorhexidine/silver sulfadiazine will develop is currently unknown. CVC impregnated with minocycline/rifampin have also been trialled21, but there is some concern that the use of catheters impregnated with minocycline and rifampicin could increase the incidence of minocycline and rifampicin resistance among pathogens, particularly staphylococci3. In vitro rifampicin resistance in staphylococci is easily induced by single step mutation after rifampicin exposure22. Rifampicin is also one of the few oral antimicrobials we have available in our armamentarium to treat MRSA infections, so it is critical to avoid unnecessary usage of this drug with the potential for induction of resistance where possible. When should i.v. catheters be changed? Scheduled replacement of peripheral venous catheters is recommended before 72 h after insertion because after this time the incidence of thrombophlebitis and bacterial colonization increases exponentially23. Obviously catheters that are visibly phlebitic or are not required clinically should be removed at once. Specialized ‘i.v. teams’ in hospitals have been shown to be unequivocally effective in reducing the incidence of peripheral catheter-related infections and the associated complications and costs24,25. When should CVC be replaced? In the past many were routinely replaced every 7 days, particularly in high acuity areas such as intensive care units (ICU), where line breaks are frequent. However, one study involving 112 surgical ICU patients showed no difference in catheter-related bacteraemia rates in patients undergoing scheduled catheter replacement at 7 days, compared to replacement on an as needs basis26. In addition, CVC are somewhat different to peripheral i.v. catheters because there is substantial morbidity associated with the practice of insertion. For these reasons, it is now recommended by the Centers for Diseases Control and Prevention Guidelines that routine replacement of CVC is unnecessary, provided the catheter is functioning and has no evidence of local or systemic complications3. The appropriate management of CASAB is to remove the catheter where possible, although the optimal duration of antibiotic therapy remains controversial27. S. aureus has a particular affinity for bone, joints, native and prosthetic heart valves and other prosthetic material. In one study of catheter-related S. aureus bacteraemia, endocarditis was noted on transoesophageal echocardiography (TOE) in 16/69 (23%) patients, even in the absence of other cardiac risk factors28. In the study by Thomas and Morris, 15/250 (6%) surviving patients with CASAB had metastatic seeding of infection, predominantly heart valve, bone or deep endovascular infection2. For this reason the treatment of S. aureus bacteraemia is more aggressive and of longer duration than bacteraemias caused by most other organisms, and the culture of S. aureus from blood is always considered clinically significant. In one study, infectious disease physicians uniformly recommended much longer durations of therapy for S. aureus bacteraemia than other medical staff29, presumably because they frequently identify and treat the peripheral metastatic foci of staphylococcal infection, which often present later on. When one considers optimal management of these infections, it is important to remove the catheter where possible. In a study of 244 hospitalized patients with S. aureus bacteraemia, patients whose intravascular device was not removed were 6.5-fold more likely to die or to relapse after controlling for other factors29. Assuming rapid removal of the offending catheter and rapid resolution of symptoms and signs of sepsis, 10–15 days of parenteral antibiotics is equivalent to longer durations of therapy30. In a meta-analysis of 25 studies including 417 patients that reported the outcome of short-course therapy for CASAB, less than 10 days of i.v. therapy was associated with higher rates of relapse31. In the study of CASAB by Thomas and Morris in this issue2, patients who responded rapidly to therapy were arbitrarily assigned to 7, 10 or 14 days of treatment duration. There was no relationship between treatment duration and risk of relapse, except that patients receiving longer than 14 days of treatment were more likely to have a deep-seated focus of infection and, therefore, received longer treatment duration. As the authors point out, their study was purely observational and there is still a need for a prospective randomised controlled trial to determine the optimal duration of therapy in CASAB that responds rapidly to removal of the catheter. Because of this, the most recent guidelines published by the Infectious Diseases Society of America still recommend a minimum of 14 days parenteral therapy27. CASAB that does not settle rapidly with the removal of the catheter should be treated for at least 4 weeks with parenteral antibiotics, and investigation for metastatic seeding including echocardiography is mandatory27. It should also be remembered that the sensitivity of transthoracic echocardiography (TTE) in the diagnosis of S. aureus endocarditis is low, with only 27% of patients in one study with endocarditis based on Duke criteria and positive TOE having endocarditis noted on TTE28. Should endocarditis be suspected, patients with a negative TTE should go on to have a TOE, and if endocarditis is confirmed the duration of parenteral therapy should be at least 6 weeks27. CASAB is a significant cause of increased morbidity, mortality and excess cost in hospitalized patients. Many of these infections can be prevented by careful attention to hand hygiene, skin antisepsis, regular inspection of the i.v., rapid removal of cannulae that are phlebitic or are not required and replacement of peripheral i.v. cannulae within 72 h of insertion. Specialized hospital ‘i.v. teams’ markedly reduce the rate of CASAB. If CASAB occurs, catheters should be removed where possible and parenteral antibiotic therapy is required. The optimal duration of parenteral therapy for CASAB that responds rapidly to the removal of the catheter is yet to be determined, but based on current evidence should be 14 days. Shorter durations of therapy may be possible in the future, but a randomised controlled trial is required to assess this before shorter durations can be recommended.
No takes yet. Share an insight, caveat, or question.
Wendy Munckhof (2005) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: