New method for assessing hand disinfection shows that pre-operative alcohol/chlorhexidine rub is as effective as a traditional surgical scrub.

BACKGROUND: Several studies have shown that rubbing hands with an alcohol/chlorhexidine solution provides equivalent microbial decontamination to a conventional surgical scrub using aqueous chlorhexidine. However, the authors believe that these studies have methodological flaws that limit their applicability to the operating theatre environment. As such, a method was developed to compare products in an everyday operating theatre environment using working operating theatre personnel. AIM: To determine whether or not an alcohol/chlorhexidine rub is as efficacious as a traditional surgical scrub using a novel method. METHODS: Bacterial counts at baseline were collected from 20 anaesthetists using the glove juice method. Subsequently, with sequential exchange of sterile gloves, one hand underwent a 3-min scrub using 4% aqueous chlorhexidine, and the other hand underwent a 60-s rub with a 70% isopropyl alcohol/0.5% chlorhexidine solution. The residual bacterial count was collected for each hand after 30 min using the glove juice method. These counts were converted to log10 values to compare the baseline counts of right and left hands, and efficacy between the treatment groups. FINDINGS: Mean [± standard deviation (SD)] bacterial counts at baseline were (log10) 4.42 ± 0.81 for left hands and 4.64 ± 0.60 for right hands (P > 0.05). The mean (± SD) reduction from baseline was (log10) 1.45 ± 0.50 for 4% chlorhexidine and 2.01 ± 0.98 for alcohol/chlorhexidine (P > 0.05). CONCLUSION: An alcohol/chlorhexidine hand rub was found to be as efficacious as a traditional scrub after 30 min; this study differs from previous work as it was undertaken in a population of practising anaesthetists in their working environment. The McKenzie method allows baseline and study evaluations to be performed contemporaneously on the same individual. Each subject was his/her own control. This method offers a more clinically relevant way to compare disinfectant solutions than standard methods.

Successful treatment of iatrogenic multicentric Castleman's disease arising due to recrudescence of HHV-8 in a liver transplant patient.

We describe the case of a 59-year-old HIV-negative male who developed multicentric Castleman's disease (MCD) 1 year postliver transplantation due to recrudescence of a pretransplant human herpesvirus-8 (HHV-8) infection. He presented with fevers, dry cough, weight loss and drenching night sweats. Routine investigations were all unremarkable. Computerized axial tomography (CT) scans showed splenomegaly and intra-abdominal lymphadenopathy, confirmed by positron emission tomography. Cervical lymph node biopsies were consistent with MCD. The presence of HHV-8 was confirmed on immunohistochemistry. Peripheral blood HHV-8 quantitative polymerase chain reaction (qPCR) monitoring showed a threefold decrease in viremia in the first week of treatment with ganciclovir but had little impact on clinical symptoms. Reducing immunosuppression and switching to rituximab resolved clinical symptoms and produced a negative HHV-8 qPCR result. Retrospective molecular testing of sera collected pre- and immediately posttransplantation confirmed preexisting HHV-8 in the host. This is the first reported case of an HIV-negative postliver transplant patient developing MCD that manifested as posttransplant lymphoproliferative disorder due to recrudescence of HHV-8. We propose (1) the introduction of the term iatrogenic Castleman's disease (CD) for this and similar cases, (2) rituximab should be considered as a treatment option for CD and (3) consideration be given to a change to the World Health Organization classification of CD to incorporate such cases.

Catheter-associated bloodstream infection incidence and risk factors in adults with cancer: a prospective cohort study.

Central venous catheter-associated bloodstream infections (CABSIs) cause considerable morbidity in patients with cancer. We determined the incidence and risk factors for CABSI by performing a prospective observational cohort study of all adult patients requiring a central venous access device (CVAD) in a haematology-oncology unit. All CVADs were inserted under ultrasound guidance by trained operators in a dedicated interventional radiology facility. A total of 1127 CVADs were assessed in 727 patients over 51,514 line-days. The rate of CABSI per 1000 line-days was 2.50. Factors associated with CABSI included: type of CVAD, greatest for non-tunnelled lines [hazard ratio (HR): 3.50; P < 0.0001] and tunnelled lines (HR: 1.77; P = 0.011) compared to peripherally inserted central venous catheter (PICC) lines; patient diagnosis, greatest for aggressive haematological malignancies (HR: 3.17; P = 0.0007) and least for oesophageal, colon and rectal cancers (HR: 0.29; P = 0.019) compared to other solid tumours; side of insertion, greatest for right-sided lines (HR: 1.60; P = 0.027); and number of prior line insertions (HR: 1.20; P = 0.022). In patients with aggressive haematological malignancies there was significantly more CABSI with non-tunnelled lines (HR: 3.9; P < 0.001) and a trend to more CABSI with tunnelled lines (HR: 1.43; P = 0.12) compared to patients with PICC lines, as well as increased CABSI for right-sided insertions (HR: 1.62; P = 0.047). This study highlights the utility of a standardised CABSI surveillance strategy in adult patients with cancer, provides further data to support the use of PICC lines in such patient populations, and suggests that the side of line insertion may influence risk of CABSI.

Genetic relatedness and virulence gene profiles of Escherichia coli strains isolated from septicaemic and uroseptic patients.

We investigated the relationship between clonality and virulence factors (VFs) of a collection of Escherichia coli strains isolated from septicaemic and uroseptic patients with respect to their origin of translocation. Forty septicaemic and 30 uroseptic strains of E. coli were tested for their phylogenetic groupings, genetic relatedness using randomly amplified polymorphic DNA (RAPD), biochemical fingerprinting method (biochemical phenotypes [BPTs]), adherence to HT-29 cells and the presence of 56 E. coli VF genes. Strains belonging to phylogenetic groups B2 and D constituted 93% of all strains. Fifty-four (77%) strains belonged to two major BPT/RAPD clusters (A and B), with cluster A carrying significantly (P = 0.0099) more uroseptic strains. The degree of adhesion to HT-29 cells of uroseptic strains was significantly (P = 0.0012) greater than that of septicaemic strains. Of the 56 VF genes tested, pap genes was the only group that were found significantly (P < 0.0001) more often among uroseptic isolates. Phylogenetic group B2 contained a significantly higher number of strains carrying pap genes than those in group D. We conclude that uroseptic E. coli are clonally different from septicaemic strains, carry more pap genes and predominantly adhere more to the HT-29 cell model of the gut.

Use of elastin fibres detected in non-directed low volume bronchial lavage in ventilated ICU patients.

Elastin fibres in sputum have been described as a more sensitive marker of pulmonary necrosis than plain chest X-rays. This study aimed to determine the prevalence of elastin fibres using non-directed non-protected mini-bronchoalveolar lavage (BM-BAL) in mechanically ventilated patients in the intensive care unit. Patients admitted to the general intensive care unit of a tertiary referral hospital requiring more than 48 hours of mechanical ventilation had surveillance BM-BAL performed on admission and were then examined weekly using potassium hydroxide wet preparations for the presence of elastin fibres. All positive and a random selection of 16 negative preparations from patients with acute respiratory distress syndrome or pneumonia were fixed and examined using Weigert's staining method for elastin. Of 412 patients enrolled, 130 (32%) had pneumonia on admission, 50 (12%) developed 58 episodes of ventilator-associated pneumonia and acute respiratory distress syndrome was diagnosed in 86 patients (21%). No chest X-ray showed cavitating infiltrates. Of 985 specimens examined, only seven had elastin fibres. Elastin fibres are uncommonly found using BM-BAL in general screening, acute respiratory distress syndrome or pneumonia in the intensive care unit, the incidence too low to be a useful indicator of pulmonary necrosis.

A bronchoscopic scoring system for airway secretions--airway cellularity and microbiological validation.

There is currently no validated scoring system for quantification of airway secretions in children. A user friendly, valid scoring system of airway secretions during flexible bronchoscopy (FB) would be useful for comparative purposes in clinical medicine and research. The objective of this study was to validate our bronchoscopic secretion (BS) scoring system by examining the relationship between the amount of secretions seen at bronchoscopy with airway cellularity and microbiology. In 106 children undergoing FB, the relationship of BS grades with bronchocalveolar lavage (BAL) cellularity and infective state (bacterial and viral infections) were examined using receptor operator curves (ROC). BAL was obtained according to European Respiratory Society guidelines; first lavage for microbiology and second lavage for cellularity. Area under the ROC was significant for total cell count (TCC) and neutrophil % but not for lymphocyte %. BS grade significantly related to infection positive state (chi(trend) (2) = 5.85, P = 0.016). The area under the ROC for infection positive state versus BS grade was 0.645, 95% CI 0.527-0.763. The BS scoring system is a valid method for quantifying airway secretions in children undergoing bronchoscopy. The system related well to airway cellularity and neutrophilia, as well as to an airway infective state. However, the system is only complementary to cell counts and cultures and cannot replace these laboratory quantification techniques.

Airway cellularity, lipid laden macrophages and microbiology of gastric juice and airways in children with reflux oesophagitis.

BACKGROUND: Gastroesophageal reflux disease (GORD) can cause respiratory disease in children from recurrent aspiration of gastric contents. GORD can be defined in several ways and one of the most common method is presence of reflux oesophagitis. In children with GORD and respiratory disease, airway neutrophilia has been described. However, there are no prospective studies that have examined airway cellularity in children with GORD but without respiratory disease. The aims of the study were to compare (1) BAL cellularity and lipid laden macrophage index (LLMI) and, (2) microbiology of BAL and gastric juices of children with GORD (G+) to those without (G-). METHODS: In 150 children aged < 14-years, gastric aspirates and bronchoscopic airway lavage (BAL) were obtained during elective flexible upper endoscopy. GORD was defined as presence of reflux oesophagitis on distal oesophageal biopsies. RESULTS: BAL neutrophil% in G- group (n = 63) was marginally but significantly higher than that in the G+ group (n = 77), (median of 7.5 and 5 respectively, p = 0.002). Lipid laden macrophage index (LLMI), BAL percentages of lymphocyte, eosinophil and macrophage were similar between groups. Viral studies were negative in all, bacterial cultures positive in 20.7% of BALs and in 5.3% of gastric aspirates. BAL cultures did not reflect gastric aspirate cultures in all but one child. CONCLUSION: In children without respiratory disease, GORD defined by presence of reflux oesophagitis, is not associated with BAL cellular profile or LLMI abnormality. Abnormal microbiology of the airways, when present, is not related to reflux oesophagitis and does not reflect that of gastric juices.

Telepathology.

The development of a Queensland-wide videoconferencing network provided an opportunity to develop telepathology. In 1999, weekly videoconferences began with remote laboratories and clinical staff in four peripheral hospitals and the Royal Brisbane Hospital and in 2000 biweekly videoconference pathology grand rounds started across Queensland with up to six sites, from Cairns to the Gold Coast, joining in or presenting. The average number of sites connected was 3.0 in 1998, 3.5 in 1999, 4.4 in 2000 and 4.5 in 2001. Problems included the complexity of the system, timing and need for bookings, coordination of presenters and presentations, and the time needed to organize sessions, set up linkages, advertise sessions and attend the telepathology conference. Successful meetings have been associated with well prepared cases, time for discussion, attendance by all sites, timeliness of cases and responses, and the presence of experts to respond to questions, as well as effective linkages and trouble-free hardware. Future needs include better infrastructure and trained staff to coordinate the linkages and presentations. Telepathology has an important part to play in the provision of cost-effective medical care in Queensland.

Hepatitis G virus genotypes in Australia, Papua New Guinea and the Solomon Islands: a possible new Pacific type identified.

BACKGROUND: Hepatitis G Virus (HGV)/GB Virus-C (GBV-C) is a newly discovered RNA virus. Nucleotide sequence comparison and phylogenetic studies of the 5' untranslated region (5'UTR) within the viral genome have identified at least three different types which have provisionally been classified as type 1 (West African origin), type 2 (North American origin) and type 3 (Asian origin). METHODS AND RESULTS: The products of RT-PCR were sequenced by using blood donors and patients infected with HGV/GBV-C in Australia, Papua New Guinea and the Solomon Islands to investigate the genotype distribution in this area of the world. All the Australian isolates showed strong sequence homology with type 2, while the Papua New Guinea and Solomon Islands sequences were more closely related, but differ from type 3, which has previously been reported from isolates studied within Asia. CONCLUSIONS: Phylogenetic analysis suggests that these latter sequences are either a new HGV/GBV-C Pacific type or a subtype of the Asian type RNA virus. Isolates homologous with type 1 were not identified in these population groups.

Isolation of beta-lactamase positive vancomycin resistant Enterococcus faecalis; first case in Australia.

An increasing number of clinical isolates of vancomycin resistant enterococci (VRE) have been reported in the literature since 1988. Only a few cases of beta-lactamase producing VRE have been described worldwide. This article reports the first case of beta-lactamase positive VRE in Australia. This strain of Enterococcus faecalis was isolated from a patient with non-Hodgkin's lymphoma who subsequently underwent a bone marrow transplant. Screening of all ward contacts did not detect any further case of beta-lactamase producing VRE. With the development of multiple antibiotic resistance in enterococci, additional infection surveillance protocols have been implemented in the hospital. These include routine screening of 'at risk' patients, instigating relevant infection control measures for management of VRE positive patients and controlling the usage of vancomycin in order to limit the development of resistant isolates.

Successful treatment of post-influenza pseudomembranous necrotising bronchial aspergillosis with liposomal amphotericin, inhaled amphotericin B, gamma interferon and GM-CSF.

A case of aspergillus tracheobronchitis following influenza A infection in an immunocompetent 35 year old woman is described that required prolonged mechanical ventilation for airways obstruction. Treatment included liposomal amphotericin, inhaled amphotericin, gamma interferon and GM-CSF. Liposomal amphotericin therapy was associated with reversible hepatosplenomegaly. Inhaled corticosteroids with continued antifungal therapy were used for the management of severe recurrent airway obstruction. After a prolonged course of treatment she survived with fixed airways obstruction unresponsive to corticosteroids.

Comparison of the antibacterial efficacy of 4% chlorhexidine gluconate and 1% triclosan handwash products in an acute clinical ward.

The antibacterial efficacy of 4% chlorhexidine gluconate (CHG) and 1% triclosan as handwash antiseptics is well established. Few published studies have identified hand bacteria found in glove juice samples, and most studies have used nonclinical study subjects. We report a longitudinal comparative study to determine the effect of 4% CHG and 1% triclosan on the composition of the hand bacterial flora of clinical staff in a specialist surgical unit. Prehandwash and posthandwash samples were collected on 3 separate occasions throughout each day by using the glove juice method and a supervised handwashing technique. Total bacterial counts were determined as well as counts for specific pathogens including methicillin-resistant Staphylococcus aureus and coliforms. Both 4% CHG and 1% triclosan were found to effectively reduce the total hand bacterial count preduty (P =.0001). Four percent CHG also was consistently more effective at reducing the total count than was 1% triclosan. However, 1% triclosan eliminated methicillin-resistant S aureus, whereas 4% CHG failed to do so (P =.0001). Gram-negative bacteria were more likely to be eliminated after the use of 4% CHG compared with 1% triclosan. This study is the first to report the effects of 1% triclosan on the bacterial flora present on the hands of clinical staff and demonstrates the ability of 1% triclosan to eliminate methicillin-resistant S aureus.

The serological status of Solomon Island blood donors.

The serological status of Solomon Island blood donors in 1995 and in particular the seroprevalence of antibodies to Hepatitis B and C and prevalence of risk factors for these chronic infections was studied. A questionnaire of risk factors for Hepatitis B and C was undertaken. All blood donors had been previously screened for HIV antibody without any positive cases recorded. 598 donors had serum collected of which 36 samples (6.0%) were third generation HCV EIA antibody positive and 3 samples were RIBA positive but none were PCR positive. 25.1% of samples were positive for HBsAg and anti-HBc antibody was found in 84.4%. Elevated ALT levels (>35 U/l) were found in 6.5% of samples but there was no statistically significant association with HCV or HBsAg status. 15.4% were TPHA positive and 5.4% had RPR titers more than or equal to 1. Anti-HTLV-1 antibody was positive in 12.3% randomly selected samples. All 10 positive samples were then found to be antibody indeterminate with Western blot assay. Of the 585 samples with completed questionnaires, analysis of the relationship between anti-HCV status with tattoo status and ear piercing also failed to reach statistical significance. Consistent with other studies from tropical malaria-prone countries, a positive anti-HCV antibody test even by the third generation EIA is probably a false positive test in most cases. In addition, high prevalence rates of HBV, yaws or syphilis infection were demonstrated.

Burn wound excision and massive blood transfusion did not affect perioperative vancomycin levels.

INTRODUCTION: The effect of burns surgery that requires intraoperative transfusion of five or more units of blood on serum vancomycin levels was assessed. METHODS: Serum vancomycin levels were measured 10 min and 6 h after vancomycin administration with surgery being performed in the interval. The following day the same dose of vancomycin was given and serum vancomycin levels measured at the same times without intervening surgery. RESULTS: Thirteen operations involving nine patients who required a mean blood transfusion of 9.2 units were studied. There was very little difference between 10-min levels, 6-h levels and the change over interval (absolute and percentage) on the day of surgery and the following day. The recorded serum levels were often at the lower end of the desired range, especially in patients who underwent the larger operations. This was the case on both day of surgery and the control day. CONCLUSIONS: Large volume blood loss and replacement during burns surgery did not significantly affect perioperative vancomycin levels. 1998 Elsevier Science Ltd for ISBI.

Exposure to GB virus type C or hepatitis G virus in selected Australian adult and children populations.

BACKGROUND: The epidemiology and disease association for the GB virus type C (GBV-C) or hepatitis G virus (HGV) are poorly understood. STUDY DESIGN AND METHODS: This study describes the exposure rates to GBV-C/HGV in diverse Australian population groups by testing for current infection and evidence of past infection with a reverse transcriptase polymerase chain reaction and an anti-E2 enzyme-linked immunosorbent assay, respectively. Subjects included volunteer blood donors, hepatitis C antibody (anti-HCV)-positive donors, children, hemodialysis patients, pregnant women attending a prenatal clinic, injecting drug users (IVDUs), and adult hemophiliacs. RESULTS: Combined GBV-C RNA and E2 antibody prevalence was 6.5 percent (6/93) in children, 13.3 percent (75/565) in blood donors, 14 percent (14/99) in pregnant women, 22.5 percent (18/80) in hemodialysis patients, 80 percent (56/70) in anti-HCV-positive donors, 88.6 percent (31/35) in IVDUs, and 85.7 percent (54/63) in adult hemophiliacs. Children had the lowest antibody rate, 1.1 percent, whereas the rate was 10.8 percent for blood donors and rose to 45.7 percent for IVDUs, 57.1 percent for anti-HCV-positive donors, and 74.6 percent for hemophiliacs. In contrast, current infection rates were comparable for children, blood donors, and pregnant women (5.4, 2.6, and 6%, respectively), rising to 11.1 percent for hemophiliacs, 24.3 percent for anti-HCV-positive donors, and 48.6 percent for IVDUs. Ten of 12 blood donors had persistent viremia, while 2 had recent infections, 1 with apparent resolution. CONCLUSION: Exposure to GBV-C can commence at an early age, although ongoing exposure may also occur among adults with no apparent risk factors. GBV-C RNA positivity was not associated with abnormal plasma alanine aminotransferase levels among blood donors.

Clinical utility of hygroscopic heat and moisture exchangers in intensive care patients.

OBJECTIVE: To compare the degree of bacterial circuit colonization, frequency of ventilator-associated pneumonia (VAP), character of respiratory secretions, rewarming of hypothermic patients, disposable costs, and air flow resistance in intensive care patients ventilated using either a heat and moisture exchanger (HME) or hot water (HW) humidifier circuit. DESIGN: A prospective, randomized blinded trial of patients in the intensive care unit undergoing mechanical ventilation. SETTING: A metropolitan teaching hospital. PATIENTS: One hundred sixteen patients undergoing mechanical ventilation for a minimum period of 48 hrs were enrolled. INTERVENTIONS: Patients were randomized to three ventilation groups using a) an HW circuit with a 2-day circuit change (n = 41); or b) a bacterial-viral filtering HME in the circuit, with either a 2-day (n = 42); or c) a 4-day circuit change (n = 33). MEASUREMENTS AND MAIN RESULTS: Circuit colonization was assessed using quantitative culture of washings taken from the circuit tubing and semiquantitative culture of swabs from the Y connectors. Sixty-seven percent of HW circuits became contaminated compared with 12% in the two HME groups (p < .0001). Median colony counts were lower in the HME groups (p < .0001). If circuits at first circuit change were contaminated in the HW group, 89% of subsequent circuit changes became contaminated compared with 0% and 25% for the 2- and 4-day HME groups, respectively. The frequency of VAP, the time to resolution of admission hypothermia, and the volume and fluidity of secretions were similar for all groups. The resistance of the HME after 24 hrs of use was < 0.025 cm H2O/L at gas flows of 40 L/min. HME use resulted in a cost reduction of $1.48 (Australian)/day. CONCLUSIONS: Circuits with a bacterial-viral filtering HME are less readily colonized by bacteria. Contamination is a random event. Humidification technique has no influence on the frequency rate of VAP, the effectiveness of rewarming, nor the character of the respiratory secretions. Breathing resistance is generally low and disposable costs are reduced when an HME is used.

Does tea tree oil have a place in the topical treatment of burns?

Burnaid is a sorbalene-based cream containing 40 mg/g of tea tree oil and 1 mg/g of triclosan. This investigation was carried out to determine the effect of Burnaid, a commercial tea tree oil preparation, against Enterococcus faecalis (ATCC29212), Staphylococcus aureus (ATCC29213), Escherichia coli (ATCC25922), and Pseudomonas aeruginosa (ATCC27853), with the activity of the base product in the commercial preparation. The organisms were suspended in sterile saline (0.5 McFarland Standard) and inoculated onto horse blood agar (E. faecalis and S. aureus) or Mueller-Hinton agar (E. coli and P. aeruginosa). One hundred microliters of Burnaid unsterilized, Burnaid sterilized and the base product (Tinasolve) were placed in duplicate in wells cut into the agar plates. Sterility and inactivation cultures were also performed on the samples. None of the samples were found to be contaminated with bacteria prior to testing. Only S. aureus and E. coli showed zones of growth inhibition around the Burnaid and Tinasolve. Zones of growth inhibition (22 mm) were similar for the active product (Burnaid) and the base (Tinasolve). There was no activity against E. faecalis or P. aeruginosa. In view of our findings and literature indicating the cytotoxicity of tea tree oil against human fibroblasts and epithelial cells, it is recommended that this product should not be used on burn wounds.

Further observations on the immune response to recombinant hepatitis B vaccine after administration to aboriginal and Torres Strait Island children.

OBJECTIVE: To determine the prevalence of markers of hepatitis B virus (HBV) immunity and infection at 5 years of age in Aboriginal and Torres Strait Island children who were fully vaccinated in infancy, and to examine the response to a booster dose of hepatitis B vaccine in those children who had no detectable immunity despite vaccination. METHODOLOGY: A cross-sectional study of serological markers to HBV in a sample of 239 Aboriginal and Torres Strait Island children, with a mean age of 5.7 years, who were fully vaccinated in infancy. The antibody response to a booster dose of hepatitis B vaccine was determined in those children in the sample who had no markers of either immunity to HBV or infection with HBV. RESULTS: Of the 239 children, 6% (95% CI 4-10%) had been infected and, of these, four were HBV surface antigen (HBsAg) positive. Of the remaining 224 children, only 41% (95% CI 35-48%) had evidence of immunity (i.e. an antibody to HBV surface antigen (anti-HBs) level of > or = 10 miu/mL) to HBV. Of the children with no detectable immunity (i.e. anti-HBs < 10 miu/mL), 113 were followed up after receiving a booster dose of hepatitis B vaccine. Of these, 84% (95% CI 76-90%) had an anamnestic response (i.e. anti-HBs < 10 miu/mL following the booster dose). Therefore 16% (95% CI 10-24%) still had no detectable immunity following the booster dose. CONCLUSIONS: This study provides further evidence that Aboriginal and Torres Strait Island children have a suboptimal response to recombinant hepatitis B vaccine. It also indicates that a considerable number of Aboriginal and Torres Strait Island children in the study age cohort have been exposed to HBV. However, despite these concerns, this study and historical data provide strong evidence that there has been a marked reduction in the prevalence of HBV infection and carriage in previously 'high risk' Aboriginal and Torres Strait Island children since the introduction of hepatitis B vaccines. Aboriginal and Torres Strait Island children who have been fully vaccinated in infancy do not require a booster dose of hepatitis B vaccine at school entry.

Silvazine (silver sulfadiazine and chlorhexidine) activity against 200 clinical isolates.

The study was undertaken to examine the in vitro efficacy of Silvazine against micro-organisms commonly found in burn wounds. Two hundred non-replicative sequential clinical isolates were collected over a 2-month period. These comprised 50 Staphylococcus aureus (methicillin sensitive), 50 Staphylococcus aureus (methicillin resistant), 50 coagulase negative staphylococci and 50 Pseudomonas aeruginosa. As there is no standard test method, the method chosen was a pour plate overlay of micro-organisms placed on a Mueller Hinton base containing duplicate wells of 0.1 ml Silvazine. Plates were incubated at 35 degrees C for up to 48 h prior to examination. All organisms tested showed zones of growth inhibition. The mean diameters of the zones of growth inhibition were similar within genera. S. aureus 19.7 +/- 1.6 mm, MRSA 16.9 +/- 1.6 mm, P.aeruginosa 15.3 +/- 1.1 mm and coagulase negative staphylococci 20.8 +/- 2.1 mm. There was no bacterial regrowth within the zones of growth inhibition following long-term plate storage. In vitro testing of Silvazine has confirmed its efficiency against common burn wound isolates.