Journal of Clinical Pharmacy and Tnerapeufics (1995) 20, 165-166

Assessment of in-line filters to prolong the life of intravenous cannulae in cystic fibrosis patients C. Richards MPharm MPharmS, L. Millar-Jones MB BCh MRCP and M. Alfaham MB ChB MRCP

Department of Pharmacy and Cystic Fibrosis Unit (Deparfmenf of Child Health), University Hospifal of Wales, Cavdifi UK

SUMMARY

Twelve patients with cystic fibrosis received 12 courses of intravenous antibiotics, each over 10-14 days, both with and without the use of an extended-life disposable filter. The design of the trial was to replace the Venflon cannulae as they became non-patent and inserting a filter on alter- nate occasions. Thus each patient acted as herhis own control. Comparison of times during which cannulae remained patent showed a 50% improve- ment with use of a filter for 4 patients and no change for 7 patients. There was no significant difference associated with the use of a filter for the group as a whole but our small sample size excludes modest improvements.

INTRODUCTION

Cystic fibrosis (CF) patients frequently receive 10- to I4-day courses of intravenous (i.v.) antibiotics, during which many patients develop infusion phlebitis, a common complication of i.v. therapy requiring changes to sites for cannulation (I, 2). Particulate contamination has been suggested as the cause for up to two-thirds of all cases of infusion phlebitis (3). Small-volume parenteral products, especially lyo- philized antibiotics, are prone to high levels of particu- late contamination (4, 5). The use of an in-line i.v. filter to remove particulate contamination reduces the inci- dence of infusion phlebitis (6), an effect which is most marked in patients receiving antibiotics (7).

A randomized cross-over trial was carried out on 12 paediatric CF patients to determine the effective- ness of in-line filters to prolong the life of intravenous peripheral cannulae.

Correspondence: C. kchards, Department of Pharmacy, University Hospital of Wales, Cardiff CF4 4XW, UK.

PATIENTS A N D METHODS

Twelve CF patients (7 female; median age, 11.0 years; range, 8.2-16-1 years) who required i.v. anti- biotic therapy, either in hospital or at home, were studied.

For administration of antibiotics, all patients had a peripheral cannula (10 patients, 22-gauge Venflon; 2 patients, 20-gauge Venflon) inserted by a doctor in the hand or forearm. Patients were randomized into two groups, A and B. Group A started their course of i.v. therapy with an extended-life disposable filter (Pall Ultipor ELD 96LYL, Pall Biomedical) and group B started their course without a filter.

When a cannula site became non-patent, the patient would ‘cross-over’, so that a patient who had a filter previously would not have one and vice versa. All filters were replaced after 96 h. The filter was strapped in place on the patient‘s arm. An injection port (Vygon injection bung) was attached to the distal end of the filter housing and drugs injected through this via a small needle. For those without filters the injection port was attached to a short extension tube (Connecta 3-way stopcock with 10 cm extension tube) connected to the cannula.

Technique of antibiotic administration was standard for all patients. Information packages for nursing and medical staff and a patient administration booklet for those patients who received therapy at home were provided. All patients received an aminoglycoside (Gentamicin, 8 mg/kg/day, or tobramycin, 10 mg/kg/ day) and a second antibiotic (Azlocillin, 300-400 mg/ kg/day, Aztreonam, 200 mg/kg/day, or Timentin, 300 mg/kg/day), all given in three divided dosages daily.

Comparisons were made of cannula survival times, filtered and non-filtered, both for the group as a whole and for individual patients, using each patient as her/his own control.

0 1995 Blackwell Science Ltd 165

166 C. Richards et al.

Approval for the study was given by the Ethical Committee of the University Hospital of Wales. Patient acceptability of the filter was established in trial runs.

RESULTS

In the group as a whole, there was a trend towards an increase in cannula survival time when filters were used, but this was not statistically significant (mean cannula survival time f 1 SD was 120.9 f 51.9 h using filters and 87 f 44.7 without filters; P=0052; Wilcoxon test). In six cases (patient nos 3-5, 9, 10, 12) the use of a filter did not prolong cannula survival, whereas in five patients (nos I, 2, 7, 8, 11) it did by at least 24 h. In 4 of these patients, use of a filter produced more than a 50% increase in survival time of the cannulae

There was no difference in the antibiotics given using cannulae with and without filters. Age, sex and weight distributions were similar in the group of patients who improved to those in the group of patients who did not.

DISCUSSION

We concluded that the use of i.v. filters may be of benefit in extending cannula life in some CF patients but that this manoeuvre did not produce a significant improvement in our group as a whole. We did note, however, that none of the patients showed a reduction in cannula survival time with the use of a filter.

Infusion phlebitis occurs in response to many physi- cochemical factors, such as cannula material (8), pH of an infusate (S), specific drugs, and particulate irritation of the venous endothelium. Failure to show an overall significant improvement in cannula survival time when in-line i.v. filters were used may have been due to this multifactorial aetiology and the small sample size used.

In line with current policies, we have now changed our practice of inserting peripheral Venflons for the administration of i.v. antibiotics for CF patients and now use peripherally inserted long lines with a heparin lock. These long lines, in comparison with short lines, show a markedly prolonged survival time and a reduced incidence of phlebitis (10).

REFERENCES

I.

2.

3.

4.

5.

6.

7.

8.

9.

10.

Brown GA (1970) Infusion thrombophlebitis. British Journal of Clinical Practice, 24, 197-200. Ryan PB, Rapp RP, DeLuca PP, et al. (1973) In-line final hltration: a method of minimising the risk of bacterial fungal and particulate contamination in intravenous therapy. Bulletin of the Parenteral Drug Association, 27,

Falchuk KH, Peterson L, McNeil BJ (1985) Microparticulate-induced phlebitis: its prevention by in-line filtration. New England Journal of Medicine, 3 12, 78-82. Backhouse CM, Ball PR, Booth S, e f a/. (1987) Particu- late contaminants of intravenous medications and infusions. Journal of Pharmacy and Pharmacology, 39,

Rebagay T, Rapp R, Bivins B, et al. (1976) Residues in antibiotic preparations. i: Scanning electron microscopic studies of surface topography. American Journal of Hospital Pharmacy, 33, 433-443. Bivins BA, Rapp RP, DeLuca PP, ef al. (1979) Final inline filtration: a means of decreasing the incidence of lnfusion phlebitis. Surgery, 85, 388-394. Allcutt DA, Lort D, McCollum CN (1983) Final inline filtration for intravenous infusions: a prospective hospital study. British Journal of Surgery, 70, 111-113. Thomas ET, Evers W, Racz GB (1970) Postinfusion phlebitis. Anesthesia and Analgesia, 49, 150-159. Marshall L, Lloyd G (1987) Intravenous fluid filtration. Care of the Critically Ill, 3, 10-17. Roberts GW, Holmes MD, Staugas REM, et al. (1994). Peripheral intravenous line survival and phlebitis prevention in patients receiving intravenous antibiotics: heparidhydrocortisone versus in-line filters. Annals of Phamacofherapy, 28, 11-16,

1-13.

241-245.

0 1995 Blackwell Science Ltd, Journal of Clinical Pharmacy 6 7%erapeutics, 20, 165-166