Your browser is no longer supported

For the best possible experience using our website we recommend you upgrade to a newer version or another browser.

Your browser appears to have cookies disabled. For the best experience of this website, please enable cookies in your browser

We'll assume we have your consent to use cookies, for example so you won't need to log in each time you visit our site.
Learn more

The care and maintenance of peripheral intravenous catheters

  • Comment

VOL: 103, ISSUE: 12, PAGE NO: 30-31

Jacqueline Randle, PhD, MSc, BA, PGDip, RN, RNT, PGCert

Associate professor, Faculty of Medicine and Health Sciences, University of Nottingham; Mitch Clarke, BA, RN, is clinical nurse specialist, Infection Control, Nottingham University Hospitals NHS Trust, Queens Medical Centre Campus, Nottingham

Abstract: Randle, J., Clarke, M. (2007) The care and maintenance of peripheral intravenous catheters.

Infection prevention and control measures have been recognised as effective in minimising the risk of infection from peripheral intravenous cannulas. However, this relies on health professionals’ compliance with guidelines for the care of patients with intravenous catheters and at times it may be that practice is inconsistent with guidelines. This article uses a reflective approach to a practice incident and examines the main issues relating to the care of cannulas, specifically discussing the evidence relating to the amount of time that peripheral cannula should remain in place.

The prevention of healthcare-associated infections (HAIs) is one of the major challenges facing the NHS (Department of Health, 2001). The DH (2005) has called for all health professionals and other NHS staff to help reduce the risk of infection in everyday practice and numerous policy documents, including Getting Ahead of the Curve (DH, 2002) and A Matron’s Charter (DH, 2004), have raised the profile of infection control. Winning Ways: Working Together to Reduce Healthcare Associated Infection inEngland (DH, 2003) identified the challenges facing health professionals in their attempts to bring and maintain infection control measures into mainstream service improvement. The implication is that it is now ‘everybody’s business’ to be actively involved in reducing HAIs, while having effective infrastructures provided by management and clinical leaders to support staff (Vaughan and Randle, 2005).

In order to ensure best-practice in infection prevention and control, based on the systematic review of research and other evidence, guidelines should be easily accessible to staff (DH, 2001). This is part of the framework for clinical governance where NHS staff are striving to continually improve the quality of service provision and promote high standards of care (DH, 2001). Furthermore, from a professional perspective, the Code of Professional Conduct (NMC, 2004) states that individual nurses have a responsibility to deliver evidence-based care. This means patients have the right to receive a uniformly high standard of care, regardless of who they are and where they are treated (DH, 2000).

A way of ensuring all patients benefit from safe and appropriate care is by the production, implementation, auditing and regular updating of clinical standards, which reflect current research findings (RCN, 2002). Using audit to monitor clinical effectiveness is in the professional competencies for infection-control nurses (Infection Control Nurses Association, 2000). King (2005) stated that these are considered as standards to work towards in order to achieve consistency in the infection prevention and control specialty.

Clinical scenario

As part of the local clinical governance strategy, we audited a clinical area in a large teaching hospital in England. A component of this audit was the insertion and maintenance of peripheral intravenous catheters/cannulas (PICs). In order to critically examine current practice and compare it with the evidence base relating to PIC care, a reflective framework based on Gibbs’ work (1988) was used (Box 1). Reflective practice is considered a route to articulate knowledge locked in practice and resolves the theory-practice gap (Freshwater, 2002). By drawing on personal experience using a reflective framework it is supposed that reflexivity can move us to greater understanding (King, 1996). Schön (1983) stated that reflective practice can provide a structure where an individual can reframe practice situations, then convert new knowledge into behaviour change.

Box1. Reflective cycle (adapted from Gibbs, 1988) 

Description: What happened?

Feelings: What were you thinking and feeling?

Evaluation: What was good and bad about the experience?

Analysis: What sense can you make of the situation?

Conclusion: What else could you have done?

Action plan: If it arose again what would you do?

Description: What happened? etc 


The audit was conducted on a ward specialising in the care of people with diabetes. Part of the audit assesses clinical practice in relation to PICs and we rated the audit resultsagainst the local trust’s standard (Box 2).

Box2. Local trust’s standard and criteria for best practice 







To ensure that IV cannulas and component parts are inserted and maintained in line with best practice and guidance and to reduce the risk of cross-infection.

The date the cannula was inserted is documented in nursing or medical notes and on the patient’s drug chart

The cannula has been in situ for less than 72 hours

The cannula site is clear of signs of infection


In order to review current practice, five patients who had PICs in situ were randomly chosen and their care was audited. One of these patients, Mr Jones, had three PICs in one forearm, two of which showed signs of phlebitis in that they were red, inflamed and palpable. As with all of the patients randomly selected, there was no documentation about when the PICs were inserted and it was evident from Mr Jones’ medical and nursing notes that there was no reason why the PICs remained in place. Mr Jones was unsure when and why the PICs had been inserted. After talking to a nurse, we were informed that the PICs were inserted in the emergency department as Mr Jones was very unwell on admission. However, the nurse was not aware of when this was.


It was recognised that the issues specific to the care of Mr Jones are common in national and international healthcare settings and not unique to this particular clinical setting. However, concern for Mr Jones was uppermost - although he had been very ill and had initially required intensive medical and nursing care, his recovery was now progressing well and it was questionable whether the PICs were still necessary. This concern was heightened as people with diabetes are at a greater risk of phlebitis than the general population (Maki and Ringer, 1991; Messner and Pinker, 1992).


As a consequence of the audit it became apparent that a discrepancy existed between practice and trust guidelines. This showed the importance of audit, in which deficiencies are identified, documented and evaluated, forming the basis of an action plan to improve performance (Lamb, 1999). Consequently, feedback was provided (both verbally and in writing) and arrangements for future meetings were made in order to develop and implement the action plan.

Many issues exist that relate to this practice scenario, including the alertness of the patient, teamworking and the education and training of staff (Catney et al, 2001). Box 3 summarises the specific issues relating to the maintenance of PICs. The RCN standards for practice (RCN, 2003), on which the trust’s standards and audit criteria are based, are also identified and practice is compared with the evidence base.

Box3. Comparison between PIC maintenance theory and practice 



Any PIC may be removed by the nurse in accordance with established organisational policies and procedures, provided they have the appropriate experience, knowledge and skills (Dolan and Dougherty, 2000)


PICs that were no longer required remained in situ 

If a catheter-related complication is suspected, the catheter should be removed immediately (Drewett, 2000)

PIC sites were showing signs of phlebitis 

The nurse should be responsible for the care and monitoring of the site, implementing interventions as necessary and documenting observations and actions in the patient’s nursing and medical notes. A peripheral cannula should be removed every 72-96 hours (depending on the type of therapy) or sooner if complications are suspected (Centers for Disease Control and Prevention, 2002; Carlson, 2001; Catney et al, 2001; Bregenzer et al, 1998; Holmes, 1998)

Documentation relating to the insertion and maintenance of PICs was absent

A PIC inserted in an emergency situation where aseptic technique has been compromised should be replaced within 24 hours

Three PICs inserted in emergency situations remained in situ (this was relayed verbally) 


The longer the patient has a PIC in situ, the greater the opportunity for micro-organisms to multiply (Curran et al, 2000). This is because medical devices offer an easy way for bacteria to spread into a patient’s tissues (Parliamentary Office of Science and Technology, 2005) as they are foreign objects that break the skin.

Although incidence of bacteraemia from PICs is low (Maki and Ringer, 1991; CDC, 2002), if it does occur it may be life-threatening, especially in immunocompromised patients. Most cases of Staphylococcus epidermidis bacteraemia are associated with infected intravascular devices (Kennedy et al, 2000; Cadorna and Watanakunakorn, 1995) and for a minority of patients PICs have the potential to cause harm due to the extracellular slime that S. epidermidis produces, which adheres to PICs then multiplies and penetrates tissue or invades cells (Public Health Laboratory Service, 2000). It is therefore recommended that PICs are removed immediately when they are medically no longer needed (Joanna Briggs Institute, 1998; CDC, 2002; RCN, 2003).

A more common catheter-related infection is phlebitis, which is largely a physicochemical or mechanical phenomenon rather than being infectious (Curran et al, 2000). When phlebitis occurs there may be an increased risk of developing a local catheter-related infection (JBI, 1998). The overall incidence for phlebitis ranges from 2.3% to 43% (Maki and Ringer, 1991); this wide range perhaps reflects the difficulties in methodologies used to determine these rates, as well as differences in definitions of phlebitis.

Symptoms of phlebitis are tenderness, redness, pyrexia with unknown cause, and exudate (RCN, 2003). Factors that influence its incidence are: age, gender, patient diagnosis (for example diabetes) and infection at another body site (Maki and Ringer, 1991; Messner and Pinker, 1992). Additional process factors include: site preparation, frequency of administration set changes, catheter material, osmolarity and dose of the drug, diluent, the use of filters, hand hygiene, aseptic or venepuncture technique, number of entries into the system and type and size of catheter (Catney et al, 2001). Another significant factor affecting the likelihood of phlebitis is the experience of staff inserting and maintaining the PIC (CDC, 2002). The presence of an IV team has been shown to decrease rates of phlebitis and other complications (Miller et al, 1996; Tomford et al, 1984). The RCN (2005) recommends using Jackson’s (1998) phlebitis scale for assessing and acting upon phlebitis and the Infusion Nurses Society’s (2000) infiltration scale for improving clinical practice in relation to PIC care.

In contrast to practice witnessed in the audit, where PICs were unnecessarily in place, the majority of studies indicate that it is best practice to change PICs every 72-96 hours; it is on these studies that the CDC (2002), RCN (2005) and JBI (1998) guidelines are based. Maki et al (1987) evaluated the safety and cost of changing administration sets (including PICs) every 72 hours in 487 hospital patients. The prevalence of contamination was higher in 72 hours than 48 hours but was not statistically significant. It was concluded that for most patients it was safe and cost effective to change PICs every 72 hours. Later Maki and Ringer (1991) in a randomised controlled trial showed phlebitis increasing cumulatively but rates of phlebitis were not substantially different in PICs left in situ for 72 hours compared with 96 hours. Collin et al (1975) also found that the incidence of bacterial colonisation increased when catheters were left in situ for more than 72 hours. This finding was further substantiated by Lai’s (1998) work, in which the methodology employed was a prospective observational study. According to the CDC (2002) this is strongly recommended for implementation and supported by some experimental, clinical or epidemiological studies as well as a strong theoretical rationale. The sample population comprised medical, surgical and oncological adult patients (n=1,292) and surveillance occurred over a one-month period in a US hospital.

When comparing practice with the evidence base previously identified, it seems apparent that a discrepancy exists. However, other studies have indicated that in specific circumstances PICs can remain in situ without any adverse effects. For example, in an overseas study, Bregenzer et al (1998) found that cannula-related complications such as phlebitis did not increase during the six days from when a cannula was sited. A more methodologically robust multicentre study involved 37 centres in the UK, Ireland and Sweden, all of which participated in a surveillance project (Curran et al, 2000). Data was collected from 2,934 PICs that had been in situ for more than 24 hours and included information on age and gender of patient, location of centre, lumen size, duration of use, what the catheter was used for, whether a pump was used, where the solution was made up and the PIC’s condition on removal, the amount of inflammation and the type and amount of discharge. After analysis of this data the results were fed back to staff and two weeks later the study was repeated, with the results again being fed back to staff. Notable factors included the duration of a PIC being in situ - there was a high statistically significant increase in the phlebitis rate with each day of device use, up to five days. After this period of time the increase was not statistically significant.

In a non-experimental study conducted in the US, Catney et al (2001) indicated that PICs can remain in situ for more than 72 hours if risk factors are evaluated. They concluded that PICs could safely remain in situ but this would require careful consideration on an individual patient basis. This study was conducted with male surgical or medical patients taken from a convenience sample of 411 patients who were receiving IV fluids as part of their medical treatment. The risk factors identified substantiated earlier work by Maki and Ringer (1991) and include: the type of administered medication; the status of the patient in terms of alertness and willingness to be taught to self-evaluate site for signs of phlebitis; and the absence of a dedicated IV team.

Although Curran et al’s (2000) and Catney et al’s (2001) studies seem to support the practice in our audit, their recommendations are for patients in specific circumstances. This does not include the patient in our scenario and from a review of the literature, it seems it is established that PICs should be removed every 72-96 hours (depending on the type of therapy) or sooner if complications are suspected (CDC, 2002; Carlson, et al, 2001; Catney, et al, 2001; Bregenzer, et al, 1998; Holmes, 1998).

However, the question of routine replacement of PICs that have been well cared for and are not showing signs of phlebitis should be reviewed (Curran et al, 2000). Curran et al (2000) suggest that the benefits of removal must outweigh the risks and discomfort of resiting. Reasons for extended dwell time include poor venous access, expected site discontinuation within 24 hours and/or the use of a saline lock.

Another difference between practice in this our scenario and theory as guided by policy and evidence is the issue of documentation. Documentation should involve complete information regarding infusion therapy and vascular access (ICNA, 2000; INS, 2000). It should therefore comply with the guidelines for records and record-keeping (NMC, 2002) and specifically identify information regarding the type, length and gauge of vascular access device product, including its name, batch and lot number. Additional documentation should include all of the following: date and time of insertion, number and location of attempts, identification of the site, type of dressing, patient’s tolerance of the insertion and the name of the person placing the device. All these should be documented in the patient’s nursing and medical notes (INS, 2000). All aspects of IV therapy should be documented according to local policy and procedures (Nicol, 1999) and should include, wherever possible, a record of the patient’s consent, which is a general legal and ethical principle (DH, 2001).

To remedy shortfalls in practice relating to infection-control measures, it is suggested that practice be firmly based on the best available evidence of effectiveness (Pratt et al, 2005). In order to move towards this, experimental or scientifically robust studies need to be conducted. The RCN (2005) recommendations relating to the scenario are graded according to level three, which is based on clinical experience and anecdote. This is not to be dismissive of the studies on which the RCN guidelines are based as all standards have been through the review process by clinical experts before they being published, however, it should be recognised that the recommendations from this report - although reiterating the recommendations by CDC and JBI - are not scientifically robust.


Infection prevention and control practice is an area that has been poorly researched and has gone virtually unchallenged (Joint Commission on Accreditation of Health Care Organisations, 1996). Bowell (1992) concludes that there is an overall lack of understanding of the components that make up informed infection-control care, which is attributed to poor knowledge, a lack of motivation and low awareness of the need for infection control (Mokabel et al, 1998). It could be that this is a contributory factor to the practice witnessed in this reflective scenario. However, recognising this, it seems that for the majority of patients, PICs should be removed every 72-96 hours and all aspects of PIC care should be appropriately documented.

Action plan

The INS (2000) suggested that research should be employed to expand the base of nursing knowledge about infusion therapy, to validate and improve practice, to advance professional accountability and to enhance decision-making. To achieve this, nurses should actively participate in research activities about infusion therapy that are relevant to their job responsibilities, education, experience and practice setting (INS, 2000). Additionally, Larson et al (1988) identified the importance of infection-control personnel understanding what factors influence compliance and how compliance can best be achieved.

This audit highlighted the difference between actual and recommended practice. It is apparent that PICs can act as a source of HAIs and health professionals should be aware of this. The RCN’s (2005) standards for infusion make clear the actions necessary to reduce HAIs associated with PICs and recommend specific care as identified above, alongside good infection-control precautions such as hand hygiene. In an attempt to improve compliance, ongoing educational programmes and regular audit and benchmarking that reflect national, professional standards of practice (Stark et al, 2002; DH, 1998) should continue to be implemented. The aim of clinical audit involves taking note of what nurses do, learning from that and changing practices if necessary (Pelczar et al, 1993). This would furnish purchasers with a measurable standard in relation to quality in infection control (Brown, 1992), but this in itself should be based on well-conducted research studies that have relevance to the context of healthcare.

  • Comment

Have your say

You must sign in to make a comment

Please remember that the submission of any material is governed by our Terms and Conditions and by submitting material you confirm your agreement to these Terms and Conditions. Links may be included in your comments but HTML is not permitted.