A prospective clinical trial to evaluate the microbial barrier of a needleless connector.

Needleless connectors are being increasingly used for direct access to intravascular catheters. However, the potential for microbial contamination of these devices and subsequent infection risk is still widely debated. In this study the microbial contamination rate associated with three-way stopcock luers with standard caps attached was compared to those with Y-type extension set luers with Clearlink needleless connectors attached. Fifty patients undergoing cardiothoracic surgery who required a central venous catheter (CVC) as part of their peri- and postoperative management were studied for microbial contamination of CVC luers following 72 hrs in situ. Each patient's CVC was randomly designated to have either the three-way stopcocks with caps (control patients) or Clearlink Y-type extension sets (test patients). Prior to, and following each manipulation of the three-way stopcock luers or Clearlink devices, a 70% (v/v) isopropyl alcohol swab was used for disinfection of the connections. The microbial contamination of 393 luers, 200 with standard caps and 193 with Clearlink attached, was determined. The internal surfaces of 20 of 200 (10%) three-way stopcock luers with standard caps were contaminated with micro-organisms whereas only one of 193 (0.5%) luers with Clearlink attached was contaminated (P<0.0001). These results demonstrate that the use of the Clearlink device with a dedicated disinfection regimen reduces the internal microbial contamination rate of CVC luers compared with standard caps. The use of such needle-free devices may therefore reduce the intraluminal risk of catheter-related bloodstream infection and thereby supplement current preventive guidelines.

A randomized, prospective clinical trial to assess the potential infection risk associated with the PosiFlow needleless connector.

The microbial contamination rate of luers of central venous catheters (CVCs) with either PosiFlow needleless connectors or standard caps attached was investigated. The efficacy of 70% (v/v) isopropyl alcohol, 0.5% (w/v) chlorhexidine in gluconate 70% (v/v) isopropyl alcohol and 10% (w/v) aqueous povidone-iodine to disinfect the intravenous connections was also assessed. Seventy-seven patients undergoing cardiac surgery who required a CVC as part of their clinical management were randomly allocated either needleless connectors or standard caps. Patients were also designated to receive chlorhexidine/alcohol, isopropyl alcohol or povidone-iodine for pre-CVC insertion skin preparation and disinfection of the connections. After 72 h in situ the microbial contamination rate of 580 luers, 306 with standard caps and 274 with needleless connectors attached, was determined. The microbial contamination rate of the external compression seals of 274 needleless connectors was also assessed to compare the efficacy of the three disinfectants. The internal surfaces of 55 out of 306 (18%) luers with standard caps were contaminated with micro-organisms, whilst only 18 out of 274 (6.6%) luers with needleless connectors were contaminated (P<0.0001). Of those needleless connectors disinfected with isopropyl alcohol, 69.2% were externally contaminated with micro-organisms compared with 30.8% disinfected with chlorhexidine/alcohol (P<0.0001) and 41.6% with povidone-iodine (P<0.0001). These results suggest that the use of needleless connectors may reduce the microbial contamination rate of CVC luers compared with the standard cap. Furthermore, disinfection of needleless connectors with either chlorhexidine/alcohol or povidone-iodine significantly reduced external microbial contamination. Both these strategies may reduce the risk of catheter-related infections acquired via the intraluminal route.

A central venous catheter coated with benzalkonium chloride for the prevention of catheter-related microbial colonization.

In an attempt to overcome infections associated with central venous catheters, a new antiseptic central venous catheter coated with benzalkonium chloride on the internal and external surfaces has been developed and evaluated in a clinical trial. Patients (235) randomly received either a triple-lumen central venous catheter coated with benzalkonium chloride (117) or a polyurethane non-antiseptic catheter (118). The incidence of microbial colonization of both catheters and retained antiseptic activity of the benzalkonium chloride device following removal were determined. The benzalkonium chloride resulted in a significant reduction of the incidence of microbial colonization on both the internal and external catheter surfaces. The reduction in colonization was detected at both the intradermal (21 benzalkonium chloride catheters vs. 38 controls, P = 0.0016) and distal segments of the antiseptic-coated catheters. Following catheter removal retained activity was demonstrated in benzalkonium chloride catheters which had been in place for up to 12 days. No patients developed adverse reactions to the benzalkonium chloride catheters. The findings demonstrate that the benzalkonium chloride catheter significantly reduced the incidence of catheter-associated colonization.

A novel serological test for the diagnosis of central venous catheter-associated sepsis.

OBJECTIVES: To determine the sensitivity and specificity of a novel antibody test for the diagnosis of intravascular catheter-related infections due to coagulase-negative staphylococci. METHODS: Sixty-seven patients diagnosed as having central venous catheter (CVC)-associated sepsis based on strict clinical criteria, including positive blood cultures, were compared to 67 patients with a CVC in situ who exhibited no evidence of sepsis. An ELISA serological test based on a novel short-chain lipoteichoic acid antigen isolated from coagulase-negative staphylococci (CNS) was used to determine the patient's serological response (IgG and IgM) to CVC sepsis caused by CNS. The specificity and sensitivity of the test was determined. RESULTS: There was a significant increase in the antibody levels (IgG and IgM) to the short-chain lipoteichoic acid in patients with CVC-associated staphylococcal sepsis as compared to the control patients. CONCLUSIONS: This new serological method may offer a useful diagnostic test for intravascular catheter infections caused by staphylococci.

Use of pulsed field gel electrophoresis to determine the source of microbial contamination of central venous catheters.

Microorganisms detected in situ on the distal tip of central venous catheters (CVC) within 90 min of insertion were investigated using pulsed-field gel electrophoresis to analyse genomic fragments obtained with the SmaI restriction enzyme. Thirty patients received a triple lumen CVC, which was inserted directly through the skin using the Seldinger technique. In a further 30 patients a triple lumen CVC was inserted through a Swan sheath, thereby avoiding direct contact of the CVC with the skin. Staphylococci were isolated from the distal tips of the catheters in 6 patients (5 who had the CVC inserted directly through the skin and 1 who had the CVC inserted via a Swan sheath.) Twenty-three staphylococcal isolates were also isolated from the insertion equipment and the skin swabs surrounding the insertion site of these six patients. All the isolates were genotyped. In one of the patients the organisms isolated from the skin were identical to those on the CVC tip. In two further patients similar organisms were isolated from the insertion equipment and the patients' skin. These results, in addition to the reduced colonisation rates observed when catheters were introduced through a Swan sheath, support the hypothesis that microorganisms from the skin are impacted onto the CVC tip and the CVC insertion equipment at catheter insertion.

Novel approach to investigate a source of microbial contamination of central venous catheters.

The potential route of contamination by skin microorganisms onto the distal tip of central venous catheters during insertion was investigated. Thirty patients undergoing cardiac surgery who required a central venous catheter (CVC) as part of their clinical management were studied. Following catheter placement, the device insertion equipment and the skin at the insertion site were sampled for microorganisms. The distal tips of the CVCs were also sampled in situ within 90 min post insertion. Bacteria were isolated from 20 of 30 (66%) CVC skin insertion sites, from 15 of 30 (50%) guidewires, and from five of 30 (16%) catheter distal tips in situ. These findings suggest that despite rigorous skin disinfection and strict aseptic technique, viable microorganisms are impacted during insertion onto the distal tip of the CVC, which may act as a subsequent nidus of infection.

Surgery of the thoracic aorta with hypothermic circulatory arrest: experience with retrograde perfusion via the superior vena cava and demonstration of cerebral perfusion.

OBJECTIVE: Retrograde cerebral perfusion (RCP) via the superior vena cava has been described as an adjunctive technique to enhance the safety of hypothermic circulatory arrest (HCA), but perfusion of cerebral tissue in humans during RCP has not been demonstrated to date. We report our clinical experience with RCP and our attempt to demonstrate "true" perfusion of the brain. METHODS: Between April 1993 and June 1995, 49 thoracic aortic procedures were performed in 48 patients (male:female = 26:22) (emergency: elective = 25:24). The indications for surgery were acute type "A" dissection (18) chronic aneurysm (28) and infected valved conduit (3). Hypothermic circulatory arrest (15 degrees C) and RCP were implemented in all cases (mean HCA time 29 min, range 11-69) (mean RCP time 26 min, range 10-65). The 99mTechnetium labelled brain perfusion agent d,l, hexamethyl propylene amine oxime (99mTc-HMPAO) was administered (100 MBq) into the cardiotomy reservoir following institution of HCA (15 degrees C) in three consecutive patients and planar dynamic brain imaging with a portable gamma camera was commenced at the start of RCP. RESULTS: Six hospital deaths (12.2%) occurred in the emergency group due to atheromatous embolic stroke in one patient, sepsis in one, ruptured infrarenal aortic aneurysm in one, myocardial failure in one, renal failure in one and multi-system organ failure in one patient. The remaining patients suffered no major neurological complications (median Intensive Care Unit stay 1 day, range 1-5). Inspection of the images acquired showed 99mTc-HMPAO activity spreading quickly from the jugular bulb and the superior sagittal sinus throughout the cerebral white and gray matter. Time-activity curves calculated for both cerebral hemispheres showed homogeneous regional cerebral perfusion. CONCLUSIONS: Retrograde cerebral perfusion is easy to establish, "safe" and provides blood flow to the brain during HCA. The flow quantification and metabolic contribution of RCP require further investigation.

An audit programme for central venous catheter-associated infections.

Central venous catheters (CVC) are commonly associated with both localized and systemic infection. This paper describes an audit programme which was designed to review clinical practice associated with the insertion and subsequent care of CVC and their possible relationship to the development of infection. The programme was produced by a joint working group of the Hospital Infection Society and the Research Unit of the Royal College of Physicians and is subdivided into an audit of hospital practices and patient management associated with the use of CVC. It consists of a number of questions designed to evaluate all areas of CVC use, including the assessment of the incidence of associated local and systemic infection. Use of the questionnaire should allow policies for CVC insertion and care to be reviewed thereby, facilitating improvements, which may in turn reduce the incidence of infection associated with CVC.

Retrograde cerebral perfusion: clinical experience in emergency and elective aortic operations.

We recently have used retrograde cerebral perfusion via the superior vena cava in association with hypothermic circulatory arrest as an adjunct to cerebral protection during aortic arch operations. Between April 1993 and March 1994, 23 patients (14 male; 9 female; median age, 64 years; age range, 25 to 76 years; 14 emergency, 9 elective) underwent operation on the ascending aorta, aortic arch, or both for acute dissection (11) or aneurysm (12). Aortic root replacement was performed in 13 patients (7 with arch replacement), ascending aortic replacement in 7 (4 with arch replacement), isolated aortic arch replacement in 2, and repair of sinus of Valsalva aneurysm in 1. Coronary artery bypass grafting was performed in 4 patients. Hypothermic circulatory arrest (15 degrees C) and retrograde cerebral perfusion were implemented in all cases (median circulatory arrest time, 21 minutes; range, 13 to 51 minutes; median retrograde cerebral perfusion time, 20 minutes; range, 12 to 50 minutes). Three hospital deaths occurred (atheromatous embolic stroke, sepsis, rupture of infrarenal aortic aneurysm). The remaining patients had no neurologic damage (median intensive therapy unit stay, 1 day; range, 1 to 5 days). Retrograde cerebral perfusion is easy to establish and safe, and may improve brain protection during hypothermic circulatory arrest.

Guidelines for good practice in central venous catheterization. Hospital Infection Society and the Research Unit of the Royal College of Physicians.

Central venous catheters (CVC) are commonly used in clinical practice and are associated with several complications, including early and late onset infection. In these guidelines, an outline of good practice for the use of CVC and the prevention of associated infections is presented. Definitions of both localized and systemic catheter-related sepsis are given. Subsequent good practice in relation to the insertion of CVC, including patient preparation, planned duration of catheterization, catheter materials and design of the CVC, are presented. Skin fixation and insertion site care, including the use of dressings and administration sets, as well as an approach to flow obstructions, are also reviewed. The clinical and microbiological diagnosis of catheter-related sepsis and its treatment is next presented. Finally, guidelines for CVC removal and replacement are given. The guidelines are designed to facilitate the development of good practice in the use of CVC, allowing appropriate protocols to be formulated and to reduce infection risk.

Infections and intravascular devices.

Complications associated with intravascular devices include infections mainly caused by Staphylococcus epidermis and S. aureus. The reported incidence of these infections varies. Several factors influence the propensity for catheter infections. We recommend strategies for the prevention and treatment of catheter-related sepsis.

Dose requirements of propofol by infusion during nitrous oxide anaesthesia in man. II: Patients premedicated with lorazepam.

The infusion rate of propofol required to supplement 67% nitrous oxide in oxygen to maintain surgical anaesthesia was determined in 72 patients premedicated with lorazepam. Following an induction dose of propofol 2 mg kg-1, groups of eight patients received an infusion of propofol varying from 60 to 200 micrograms kg-1 min-1. Probit analysis was used to determine the ED50 (130 micrograms kg-1 min-1; 95% confidence limits: 106-167 micrograms kg-1 min-1) and ED95 (348 micrograms kg-1 min-1; 95% confidence limits: 233-1296 micrograms kg-1 min-1) for propofol infusion. Whole blood propofol concentrations at the time of surgical incision correlated strongly with the infusion rate, giving an EC50 value of 2.5 micrograms ml-1, and an EC95 value of 5.92 micrograms ml-1. There was no significant correlation between the rate of infusion of propofol, or the total propofol dose, and the times to response to command, or to recall of birthdate.