The use of integral fluid filters in single-use infusion administration sets

Joanna Ford (R&D Officer, SMTL)  and Pete Phillips (Director, SMTL)

November 2008

NOTE: the research in this paper has not been updated since 2008.

Introduction

This report has been produced to investigate recommendations for the use of in-line filters in IV administration sets. It sets out to clarify whether filters are required at all and if so, what pore size is recommended.

In-line fluid filters may be used within an IV administration set in an attempt to prevent (or delay) phlebitis. Phlebitis is inflammation of the vein which can be painful and can result in fever.  It can be caused in the following ways:

• Infectious agents enter the patient's vein via the administration of fluids/medication which can be the cause of nosocomical infection
• Undiluted or undissolved drug can enter the patient and damage the vein (chemical irritation)
• Particles of plastic or steel (from administration equipment such as tubing or needles) can cause vascular damage (mechanical irritation) (Falchuk et al. 1985)

All commercially available pharmaceutical IV fluids are filtered to at least 0.45 microns at the end of the manufacturing process (communication with member of UK Parenteral society).  Fluids with heat-labile components are sterilised by filtration and all other fluids are terminally sterilised.  If a solution requires bacterial filtration during administration, the filter must have a pore size no larger than 0.2 microns (RCN guidelines, 2005).  NICE guidelines recommend that in-line filters should not be used routinely for infection control purposes during infusion through central venous catheters (NICE guidelines, 2003) although no similar recommendations could be found relating specifically to peripheral intravenous catheters.  The issue of bacterial retention filters impeding the flow of some solutions has also been raised (Dunleavy and Sevick, 2001).    

Filtration properties of IV filters

Research has shown that patients undergoing IV infusions are at far greater risk of developing infusion-related phlebitis than infusion-related infection (Roberts et al. 1994).  One laboratory study indicated that a patient is likely to be infused with more than 107 particles during a 24 hour session of IV therapy (Backhouse et al. 1987). Studies using IV filters reveal that much of the particulate matter that enters the patient's vein goes unnoticed because of its size and appearance (Stromberg and Wahlgren, 1989).  These particles include scrapings of plastic that may peel off the walls of the plastic tube and enter fluid upon injection of medication as well as precipitations within the fluid.  These filters may also remove chemical particulates that have remained undissolved in the fluid.  Filters which are used to remove particulate matter of this nature have much larger pore sizes than bacterial filters.  Pore sizes for particulate filters range from approximately 15 microns (British standard 8536) to 45 microns (Chamberland et al. 1976) 

Published studies investigating the benefits of in-line filters on the incidence of phlebitis in patients have focused mainly on bacterial filters.  One randomised study used a bacterial in-line filter (0.22 microns) and reported that patients had a 75% chance of being phlebitis-free after 3 days (sets were changed each day) whereas they had only a 42% chance if the filter was not used (Falchuk et al. 1985).  Another randomised study (Allcutt et al. 1983) found that the use of a 0.22 micron filter delayed the onset of phlebitis in patients and this was particularly marked in patients who were being administered antibiotics.  Other investigators reported significantly less cases of phlebitis when an in-line filter was used  compared to infusions with no filter (Chee and Tan, 2002).  Another study (Roberts et al. 1994) found that an in-line microbial filter (0.22 microns) was as effective as heparin/hydrocortisone administration in reducing the occurrence of phlebitis when administering IV antibiotics to patients with cystic fibrosis. However, not all studies found a difference in phlebitis rates between filter groups and controls (Maddox, 1983).

Most of the research in this area has been carried out on bacterial filters and there is very little data reviewing the benefits of filters with larger pore-sizes.  One investigation found no decrease in phlebitis rates when using a filter of a  larger pore size of 45 microns (Chamberland et al. 1976).  Another found no significant difference in phlebitis rates between a 5 micron filter group and 'no-filter' control group whereas  a difference was seen between the control group and a '0.45 micron filter' group indicating that filters with smaller pore sizes appear to be more effective in reducing phlebitis rates. (Rusho, 1979). Research indicates that IV fluids contain significantly more smaller particles (less than 2 microns) than larger particles (larger than 40 microns) but what is not clear from the literature is whether larger particles cause more damage.   

The Cochrane Collaboration (Foster et al. 2006) carried out a literature review of studies investigating whether in-line filters prevented death or morbidity in neonates with IV lines. They concluded that there was insufficient evidence to link the use of in-line filters with improved clinical outcomes in this patient group.

Standards

There are a range of standards currently used in the UK which relate to the requirement of fluid filters within administration sets for the filtration of larger particles.  The section below summarises the relevant standards.

Standard: BS 2463: Part 2:1989

Title: Transfusion equipment for medical use: Part 2: Specification for administration sets
Scope: Administration sets (for blood and other fluids).  All administration sets shall include a 'filter enclosed in a transparent chamber'.  The filter chamber may be integral with the drip chamber.
Filter specifications: Administration sets for use with IV fluid other than blood should have a pore diameter that prevents the passage of particles of a size of 40 microns or greater.

Standard: BS EN ISO 8536-4:2007

Title: Infusion equipment for medical use. Part 4: Infusion filters for use, gravity feed
Scope: Gravity feed infusion sets.  Components of solution sets include a fluid filter.
Filter specification: 'Generally fluid filter used has nominal pore size of 15 microns'

Standard: BS EN ISO 8536-8:2004

Title: Infusion equipment for medical use. Part 8: Infusion equipment for use with pressure infusion apparatus. (For use with IV pumps, pressures of up to 200 kPa)
Scope: General requirements of vented and non-vented infusion sets.  Components of infusion sets include a fluid filter.
Filter specification: 'Generally fluid filter used has nominal pore size of 15 microns'

Standard: BS EN ISO 8536-11:2004

Title: Infusion equipment for medical use. Part 11: Infusion filters for use with pressure infusion equipment
Scope: The fluid filters within pressure infusion sets
Filter specification: Includes design and performance specifications (for example, filter housing should be transparent and possess a venting system in case of blockage by air bubbles). Filter specification does not include effectiveness or separation of particles (no pore size included). 

Confusion between current standards

For many years, the British Standard (BS 2463 part 2) was the relevant UK standard related to administration sets and is still commonly referred to.  However, it was withdrawn in 2007 and the European ISO standard  (BS EN ISO 8536) was adopted in the UK.   The BS 2463 specified a filter with a pore size of 40 microns or less whereas the BS EN ISO 8536 states that, in general, filters have a 'nominal' pore size of 15 microns.

A summary of the relevant standards are shown in the table below:

Recommendations

The RCN make the following recommendations:

Use of filters should adhere to the manufacturer's guidelines and the filtration requirements of the therapy
For non-lipid-containing solutions that require filtration, an additional 0.2 micron filter containing a membrane that is both bacteria/particulate-retentive and air-eliminating should be used. All infusion sets should contain in-line filtration appropriate to the solution being administered. Clear fluids require 15 micron filtration (or less) which is usually provided by a standard clear fluid set. .
For lipid infusions or total nutrient preparations that require filtration, a 1.2 micron filter containing a membrane that is both bacteria/particulate-retentive and air-eliminating should be used. (Section 4.3, RCN 2005)

The administration set used to administer IVIG (Intravenous immunoglobulins) should have a 15 micron filter to prevent infusion of undissolved immunoglobulin or other foreign material into the patient. (Section 8.9, RCN 2005)

The RCN provide the additional guidelines for the administration of parenteral nutrition :

Parenteral nutrition solutions not containing lipids should be filtered with a 0.2 micron filter during administration, or as specified in the product information .

Parenteral nutrition solutions containing lipid emulsion should be filtered using a 1.2 micron filter during administration, or as specified in the product information.(Section 8,5 RCN, 2005)

The Centre for Disease Control and Prevention (CDC) made the following statement in a MMWR report (O'Grady et al. 2002):

'Infusate-related BSI is rare'

and they highlight the fact that some solutions  can block the filter. The authors recommend filtration by pharmacy as a 'practical and less costly way to remove the majority of particulates' . 

The Infection Control nurses association (ICNA) published 'Guidelines for preventing intravascular catheter-related infection' in 2001. They did not issue specific guidelines about filter use however, stating that they are 'commonly used to administer IV drugs of high molecular size to reduce the risk of phlebitis and accidental air administration into the circulatory system. However their use in preventing CR-BSI remains controversial' (ICNA, 2001).

Conclusion

It is clear from the literature reviewed here that filters can play a useful role in preventing and reducing phlebitis.  However, up to date clinical data on large-pore filters is sparse and further research in this area would clearly be beneficial.

Whenever infusion sets are used to administer solutions, instructions for use should be followed for both the device and the medicinal product.  If compliance with an appropriate standard is claimed on the packaging, then any requirement for a filter should be adhered to by the manufacturer.  However, as standards are voluntary, many manufacturers do not claim or test for compliance with relevant standards

The MHRA have advised SMTL (personal communication) that the decision ultimately lies with the healthcare professional, based on best practice as to whether an in-line filter is required and what pore size is appropriate for that particular therapy.

In conclusion, although there is some good clinical evidence for the role of bacterial retention filters in reducing phlebitis, and guidelines exist for clinicians in determining which filters are appropriate for certain fluids, we have been unable to find guidelines which clearly state the circumstances in which filters should be used.

Bibliography

Allcutt DA, Lort D and McCollum CN (1983) Final inline filtration for intravenous infusions: a prospective hospital study. Br J Surg v 70 p 111-113

Backhouse CM, Ball PR, Booth S, Kelshaw MA, Potter SR and CN McCollum (1987) Particulate contaminants of intravenous medications and infusions. J Pharm Pharmacol v 39 p 241-245

Chamberland ME, Lyons RW, Brock SM (1976) Effect of inline filtration of intravenous infusions on the incidence of thrombophlebitis. Am J Hosp Pharm v 34 p 1068-1070

Chee S, Tan W (2002) Reducing infusion phlebitis in Singapore hospitals using extended life end-line filters. J Infus Nurs. v25 p95-104

Dunleavy M, Sevick S (2001) Medical device technology, May p 10-15

Falchuk KH, Peterson L, McNeil BJ (1985) Microparticulate-induced phlebitis. Its prevention by in-line filtration. N Engl J Med. v 312 p78-82

Foster J, Richards R and Showell M (2006) Intravenous in-line filters for preventing morbidity and mortality in neonates (Review). Cochrane database of systematic reviews Issue 2

ICNA (2001) Guidelines for preventing intravascular catheter-related infection. ICNA publication.

Maddox R, John J, Brown L, Smith C. (1983) Effect of inline filtration on postinfusion phlebitis. Clinical Pharmacy v2 p58-61

NICE guidelines (2003) Prevention of healthcare associated infections in primary and community care. Web site www.nice.org.uk/pdf/infection_control_fullguide.pdf (accessed November 2005)

O'Grady N et al (2002) Guidelines for the prevention of intravenous catheter-related infection' MMWR v51 no. RR-10 p1-36

RCN guidelines (2005) Standards for infusion therapy. Web site www.rcn.org.uk/publications/pdf/standardsinfusiontherapy.pdf (accessed November 2005)

Roberts G W, Holmes M D, Staugas R E, Day R A, Finlay C F and Pitcher A (1994) Peripheral intravenous line survival and phlebitis prevention in patients receiving intravenous antibiotics: heparin/hydrocortisone versus in-line filters.  An Pharmacol. v 28 p 11-16

Rusho W and Bair J (1979) Effect of filtration on complications of postoperative intravenous therapy. American Journal of Hospital Pharmacy v36  p1355-1356

Stromberg C and Wahlgren J (1989) Saving money with effective in-line filters. Intensive care Nursing. v 5 p 109-113

Weinstein, SM (ed) (2001) Plumer's principles and practice of IV therapy. 7th ed. Philadelphia: Lippincott Williams and Wilkins