H Walpot, RP Franke, WG Burchard, C Agternkamp, FG Mueller, C Mittermeyer, G Kalff. Particulate contamination of intravenous solutions and drug additives during long-term intensive care. Anaesthesist 1989;38:544-548.

In this study, scanning electron microscopy (SEM), in combination with energy dispersive X-ray analysis (EDX), was used to examine lung tissue from patients that had died after long-term intensive care. Particulate material was found associated with thrombi and endothelial damage in the microcirculation.

Nine adults, one child and one baby were examined post mortem. All had died after developing irreversible respiratory distress syndrome and had been ventilated for at least three days. Cylindrical samples of tissue were collected at 12 locations in the lungs and examined using SEM and optical microscopy. EDX was used to identify the nature of material located within the tissue.

Histopathological changes consistent with respiratory distress were found in all patients. The most noticeable finding was disseminating micro-thrombosis of the microcirculation and partial macro-thrombosis. Refractory bodies were seen within microthrombi. Endothelial damage was evident. Granulocytes and macrophages were seen to contain foreign particles, these particles were found to include rubber, glass, metal and silica.

The authors conclude that five effects of particulate material were evident:

• Emboli formation in the micro-circulation
• Direct damage of the endothelium, especially the capillary endothelium
• Thrombogenesis
• Formation of granulomata
• Formation of foreign body giant cell

On the basis of these findings, the authors recommend end-line filtration of infusates in intensive care.

CJ Kirkpatrick. Microcirculatory Problems in Multiple Organ Failure-The Role of Endotoxins and Particulate Contamination. Proceedings of the Symposium-"Managing the Complications of Intravenous Therapy" 1992.

Mutiple Organ Failure (MOF) is one of the most serious disease complexes in contemporary intensive care medicine. Experimental studies, both in vivo and in vitro, as well as clinical observations indicate that from a pathophysiological point of view MOF is a problem of the microcirculation. One of the best investigated components of MOF is the adult respiratory distress syndrome (ARDS). In the latter disease it has been show that the characteristic early changes of pulmonary interstitial oedema are the result of endothelial lesions in the pulmonary microvasculature. Numerous investigations of endothelia cells in vitro have provided valuable evidence that disturbances of endothelial function may explain many of the pathogenetic steps in the development of ARDS. Among these are alteration in the balance between pro- and anti-thrombogenic functions of the endothelium, the response to endotoxins and the expression of cell adhesion molecules (CAM's), responsible for the interaction of endothelial and blood cells, such as granulocytes and monocytes.

In ARDS these pathogenetic mechanisms lead to microthrombus formation and granulocyte sequestration, which severely impair microcirculatory function. It is hypothesized that the contaminants in intravenous infusions, both of a particulate and endotoxin nature, may significantly contribute to the deleterious effects of these microcirculatory disturbances. One of the most serious of these mechanisms is the ability of foreign surfaces to activate mediator systems in plasma, resulting in microthrombus formation with the microscopic particle acting as a condensation nucleus. Evidence exists that this mechanism is also operative in vivo.

In conclusion, the data from in vivo and in vitro studies support the view that risk groups, such as intensive care patients, be protected from such contaminants by effective filtration systems, both for particulate matter and endotoxins.

RA van Lingen, W Baerts, A Marquering, GJHM Ruijs. ELD96 particle filters in sick newborn infants result in significantly fewer infectious complications at lower cost. Journal of Clinical Microbiology and Infection 1997;3:122.

This study was presented at the 8th European Congress of Microbiology and Infectious Diseases, held in Lausanne, Switzerland, in May 1997. The authors compared the rates of infection and other major complications, in patients receiving infusions with or without endotoxin-retentive filtration. The cost of disposable infusion equipment and nursing time required to change it were also compared between the two groups. Infection rates and cost were found to be reduced by the use of the filter.

88 newborn infants (76 preterm, 12 term), were randomised to receive IV fluids and medications through an endotoxin-retentive intravenous filter (Pall Posidyne ELD96) or unfiltered. In the filter group all medications and fluids except lipids, blood and blood products were filtered, and sets and filters were changed every four days. Sets were changed daily in the control group. Used filters and IV catheters, IV fluids, blood and tracheal aspirate were cultured. Phlebitis, extravasation, necrosis, thrombosis, and septicaemia were scored. Costs of infusion disposables and nursing time were monitored.

Total complications were reduced in the filter group, (9 cases versus 19). This reflected a 50% reduction in sepsis (4 cases Vs 8). The mean cost of disposables was reduced from DFl 68.64 in the control group to DFl 52.41 in the filter group. When the cost of nursing time was estimated the overall costs were reduced from DFl 188.68 in the control group to DFl 82.41 in the filter group.

The authors concluded that the use of a Posidyne ELD 96 hour filter leads to significant decreases in major complications, the cost of disposables and the nursing time required to change them.

D Cousins. Cost Savings in IV Therapy. Care of the Critically Ill 1988;4:1-4

The author reports the comparative costs of intravenous therapy in an intensive care unit after the introduction of an endotoxin-retentive IV filter which enabled the use of disposable IV equipment to be safely extended to four days. Cost savings of £8637 (about fifteen thousand US dollars) were recorded.

The costs of disposable intravenous infusion equipment during a twelve month period was assessed using a computerised stock control system located in pharmacy, this figure was divided by the number of patient bed days, giving a cost per occupied bed day. During this time, intravenous equipment was changed daily in line with national guidelines. The Pall Posidyne ELD endotoxin-retentive intravenous filter was introduced and the infusion set change interval extended from 24 to 96 hours. The costs of disposables were monitored as before, for a six month period. Nursing time involved in changing infusion equipment was estimated for the control and study period.

The cost per occupied bed day was reduced from £14.64 to £9.92 after the introduction of the filter. The savings in nursing time was estimated to be 465 hours annually, or 0.22 whole time equivalents (WTE). A reduction in drug wastage was also observed but not quantified. It was also noted that the introduction of the endotoxin-retentive filter required training of the nursing and medical staff in the correct use of filters and in the prevention of inappropriate drug mixing that produced precipitates. The presence of the filter alerted staff to these physico-chemical drug interactions.

The author concluded that the introduction of the Pall Posidyne ELD endotoxin-retentive IV filter enabled the extended use of disposable intravenous infusion equipment, which resulted in savings in costs and nursing time.

National Advisory Group on Standards and Practice Guidelines for Parenteral Nutrition. Safe Practices for Parenteral Nutrition Formulations. Journal of Parenteral & Enteral Nutrition 1998;22:49-66.

This document was approved by the board of directors of the American Society for Parenteral and Enteral Nutrition (ASPEN). It contains information and specific guidelines on the use of filtration in parenteral nutrition Section IV: In-line filtration of PN admixtures.

Three other bodies also recommend filtration in IV therapy and PN - the National Co-ordinating Committee on Large Volume Parenterals, the Intravenous Nurses Society and the FDA. The rationale for using filters is discussed, including the need to remove particulates, reduce phlebitis and prevent the infusion of precipitates. Regarding this last indication, it is noted that "Filters have sometimes been criticised because they may clog, causing infusion pumps to sound their alarm and requiring nursing intervention. It should be recognised that a clogged filter is a potential sign of a precipitate. It is never appropriate to remove a clogged filter and allow the admixture to infuse without a filter."

The role of filters in the removal of micro-organisms, endotoxins and air emboli is also reviewed, and it is noted that endotoxins are able to pass through standard membranes, but not positively charged nylon. The limitations in the use of filters are reviewed including the possibility of post-filter precipitation, flow restriction and additional cost.

There is some useful information on filter selection. The use of non-endotoxin retentive filters for more than 24 hours is warned against, and the ability of positively charged nylon to safely extend IV set life to 96 hours by safely retaining endotoxin is noted. It is noted that the safe extended IV set life may offset the additional cost of using these filters. The recommendation to use 0.2µm filters for non-lipid containing PN and a 1.2-5µ filter for mixtures with lipid is highlighted.

R Robinson & P Ball. Does the Pall TNA1E parenteral nutrition admixture filter retain Malassezia furfur? Nutrition 1998; 14:363-365.

This paper demonstrates that the Pall Lipipor TNA 1.2µm filter for lipid-containing parenteral nutrition reliably retains the neonatal fungal pathogen Malassezia furfur.

M. furfur was isolated from the skin of adult volunteers and inoculated into intravenous lipid emulsion. This was filtered through the test device at 2mL/hour to simulate a typical infusion regime used in a neonatal unit, and the filtrate cultured. Unfiltered controls were also cultured.

No growth was found in the filtrate, whereas the unfiltered controls were found to be positive for M. furfur.

The authors concluded that the Pall TNA filter reliably retained M. furfur.

SN Bennett, MM McNeil, LA Bland, MJ Arduino, E Villarino, DM Perrotta, DR Burwen, SF Welbel, DA Pegues, L Stroud, PS Zeitz, WR Jarvis. Post-operative infections traced to contamination of an intravenous anesthetic, propofol. New England Journal of Medicine 1995;333:147-154.

This reports an investigation of a series of infections in patients after surgery, in seven hospitals.

Case-control and cohort studies were used to investigate the incidents. Any patients developing an acute febrile episode or organism-specific infection after surgery in the seven hospitals involved were included. Procedures and microbiological data were reviewed.

49 out of 62 patients identified had undergone surgical procedures. Several potential risk factors were identified, but only exposure to a lipid-based anaesthetic agent, propofol, was associated with the post-operative infectious complications in all seven study hospitals. Aetiological agents were identified in six outbreaks, these were Serratia marcescens, Staphylococcus aureus, Moraxella osloensis, Candida albicans and Enterobacter agglomerans. Unopened containers of the anaesthetic agent were all sterile, but samples taken from syringes of drug in use were found to be contaminated in two hospital and the same organism was recovered from case patients in one of these hospitals. Breakdown in aseptic technique was identified from interviews and observation of anaesthetic technique.

The authors conclude that handling lipid-based agents requires strict adherence to aseptic technique, to prevent extrinsic contamination and infectious complications.

RL Nichols, JW Smith. Bacterial contamination of an anesthetic agent. New England Journal of Medicine 1995;333:184-185.

This is an editorial comment on the publication by Bennett et al, in the same issue of the journal. The authors note that post-operative infections in the USA recorded for the period 1987 to 1990 have improved over those for 1975-6. Wound infections constitute 24% of the approximately 2 million annual hospital-acquired infections. Variations between procedure, hospital and surgeon and with patient characteristics. Infection control policies and antibiotic prophylaxis decrease infection rates, and the resultant reductions in mortality and morbidity are cost effective. Prevention of nosocomial infection is even more important in view of the emergence of multiply-resistant bacteria.

It is noted that the outbreaks of nosocomial infection reported by Bennett et al were found to be associated with the use of a lipid-based anaesthetic agent that is known to support the growth of bacteria. The same pathogen was isolated from several patients in six of the seven hospitals. It was shown that the contamination was not intrinsic in the agent, but resulted from extrinsic contamination, and in some cases the pathogens involved were traced to personnel involved in handling the drug.

The authors note that another incident has been reported (B Veber et al, Severe sepsis after intravenous injection of contaminated propofol. Anesthesiology 1994;80:712-3) in which four patients developed Klebsiella pneumonia after undergoing clean surgical procedures in the same operating room in an 8-hour period.

It is noted that although post-surgical infections are most commonly attributed to the procedure or surgeon, the outbreaks described by Bennett illustrate that patients are also at risk from infection associated with the anaesthesia, as well as other sources. Since 1990, at least 155 patients in 38 clusters have been apparently infected and this is an underestimate of the actual incidence.

The requirement for proper handling of lipid-based drugs is noted, including careful adherence to the manufacturer's instructions for use. The manufacturer makes particular warning of the risk of extrinsic contamination resulting from lack of aseptic technique. The authors finally suggest that the FDA may have to require education programmes for anaesthesia personnel and consider restricting the use of this type of drug to certain controlled facilities and to single use syringes.