Regional dissemination of vancomycin-resistant enterococci resulting from interfacility transfer of colonized patients.

During early 1997, the Siouxland District Health Department (SDHD; Sioux City, IA) reported an increased incidence of vancomycin-resistant enterococcal (VRE) isolates at area health care facilities. To determine the prevalence and risk factors for colonization with VRE strains at 32 health care facilities in the SDHD region, a prevalence survey and case-control study were performed. Of 2266 patients and residents, 1934 (85%) participated, and 40 (2.1%) were positive for (gastrointestinal) VRE colonization. The prevalence of VRE isolates was significantly higher in acute care facilities (ACFs) than in long-term care facilities (LTCFs) (10/152 [6.6%] vs. 30/1782 [1.7%]; odds ratio [OR], 4.1; 95% confidence interval [CI], 1.8-9.0). LTCF case patients were significantly more likely than controls to have been inpatients at any ACF (19/30 vs. 12/66; OR, 8.0; 95% CI, 2.7-23.8). Of 40 VRE isolates, 34 (85%) were a related strain. The predominant strain was present in all 12 LTCFs that had at least 1 case patient in each facility. Soon after the introduction of VRE isolates into this region, dissemination to multiple LTCFs resulted from resident transfer from ACFs to LTCFs.

Enterobacter cloacae and Pseudomonas aeruginosa polymicrobial bloodstream infections traced to extrinsic contamination of a dextrose multidose vial.

OBJECTIVE: To identify risk factors for polymicrobial bloodstream infections (BSIs) in neonatal intensive care unit (NICU) patients during an outbreak of BSIs. DESIGN: During an outbreak of BSIs, we conducted a retrospective cohort study, assessed NICU infection control practices and patient exposure to NICU healthcare workers (HCWs), and obtained cultures of the environment and HCW hands. PATIENTS: During the period May 3 to 7, 1996, 5 infants contracted BSIs caused by both Enterobacter cloacae and Pseudomonas aeruginosa, and one infant contracted a BSI caused by E cloacae only. For each pathogen, all isolates were identical on DNA typing. RESULTS: Infants exposed to the following were more likely than nonexposed infants to have BSI: umbilical venous catheters (6/14 vs 0/7, P = .05), total parenteral nutrition given simultaneously with a dextrose/electrolyte solution (6/12 vs 0/9, P = .02), or one HCW (5/7 vs 1/13, P = .007). Neither environmental nor HCW hand cultures yielded the outbreak pathogens. Quality control cultures of intravenous solution bags were negative. CONCLUSIONS: We speculate that a dextrose multidose vial became contaminated during manipulation or needle puncture and that successive use of this contaminated vial for multiple patients may have been responsible for BSIs. Aseptic techniques must be employed when multidose vial medications are used. Single-dose vials should be used for parenteral additives whenever possible to reduce the risk of extrinsic contamination and subsequent transmission of nosocomial pathogens.

Outbreak of Acinetobacter spp. bloodstream infections in a nursery associated with contaminated aerosols and air conditioners.

BACKGROUND: Acinetobacter spp. are multidrug-resistant bacteria that grow well in water and cause infections with unexplained, increased summer prevalence. In August, 1996, eight infants acquired Acinetobacter spp. bloodstream infection (A-BSI) while in a nursery in the Bahamas; three infants died and an investigation was initiated. METHODS: A case patient was defined as any newborn in the nursery during August 6 to 13, 1996, with A-BSI. To identify risk factors for A-BSI we conducted a retrospective cohort study and performed environmental cultures and air sampling using settle plates. The genetic relatedness of environmental isolates was assessed by pulsed field gel electrophoresis. RESULTS: Of 33 patients in the nursery 8 (24%) met the case definition. Patients with peripheral iv catheters were more likely to develop A-BSI (8 of 21 vs. O of 10, P < 0.05). Multivariate analysis among patients with iv catheters indicated that only exposure to one nurse was an independent risk factor for developing A-BSI (P < 0.005). Nursery settle plates were more likely to grow Acinetobacter spp. than were settle plates from other hospital areas (8 of 9 vs. 0 of 5, P < 0.005); cultures from nursery air conditioners also grew Acinetobacter spp. Environmental isolates were genetically diverse. After installation of a new air conditioner in May, 1995, A-BSIs occurred more frequently during months of increased absolute humidity or environmental dew point. CONCLUSIONS: Acinetobacter spp. may cause nosocomial BSI and death among infants during periods of polyclonal airborne dissemination; breaks in aseptic technique during i.v. medication administration may facilitate transmission from the environment to the patient. Environmental conditions that increase air conditioner condensate may predispose to airborne dissemination via contaminated aerosols and increase the risk of nosocomial A-BSI.

An epidemic of Malassezia pachydermatis in an intensive care nursery associated with colonization of health care workers' pet dogs.

BACKGROUND: Malassezia species are lipophilic yeasts that are emerging as nosocomial pathogens, particularly in low-birth-weight neonates who receive lipid emulsions. When a cluster of patients with Malassezia pachydermatis infection was identified in an intensive care nursery, we initiated an investigation. METHODS: A case patient was defined as any infant in the intensive care nursery who had a positive culture for M. pachydermatis between October 17, 1993, and January 18, 1995. We conducted a cohort study to identify risk factors for colonization and infection with M. pachydermatis. We collected cultures from the infants and the health care workers and from the health care workers' pets, since this organism has been associated with otitis externa in dogs. RESULTS: Fifteen infants met the case definition: eight with bloodstream infections, two with urinary tract infections, one with meningitis, and four with asymptomatic colonization. The case patients were significantly more likely than the other infants to weigh 1300 g or less (15 of 65 vs. 0 of 419, P<0.001). In a multivariate analysis of infants weighing 1300 g or less, the independent risk factors for colonization or infection with M. pachydermatis were a greater severity of concomitant illness (odds ratio, 19.7; P=0.001), arterial catheterization for nine or more days (odds ratio, 29.5; P=0.027), and exposure to Nurse A (odds ratio, 74.7; P=0.004). In a point-prevalence survey, 9 additional infants, 1 health care worker, and 12 of the health care workers' pet dogs had positive cultures for M. pachydermatis. The isolates from all 15 case patients, the 9 additional colonized infants, 1 health care worker, and 3 of the 12 dogs had identical patterns of restriction-fragment-length polymorphisms. CONCLUSIONS: In this outbreak, it is likely that M. pachydermatis was introduced into the intensive care nursery on health care workers' hands after being colonized from pet dogs at home. The organism persisted in the nursery through patient-to-patient transmission.

Microbiologic evaluation of needleless and needle-access devices.

OBJECTIVE: This study was carried out to determine whether needleless intravenous access devices are more likely to allow microorganisms to enter the fluid pathway than intravenous needle-access devices. METHODS: A laboratory study was conducted with two needleless and one intravenous needle-access devices and Enterococcus faecium as a bacterial challenge. Inocula of E. faecium were prepared on the basis of the numerical estimates of 1000 to 10,000 colony-forming units (CFU)/cm2 of bacterial flora on dry regions of skin (arms, legs, and hands). The septum of each access device was inoculated with 10 to 20 microliters of a 10(4) to 10(5) CFU/ml challenge suspension, which was allowed to dry on the surface of the septum. In the first part of the experiment, the needleless or needle-access cannula of each device was used to puncture the corresponding septum without previously disinfecting the top of the septum. In the second part, the contaminated septum was punctured after disinfecting the septum with a 70% isopropyl alcohol wipe. After each puncture, trypticase soy broth was flushed through the fluid pathway of the intravenous access device, collected, and cultured by the membrane filtration technique. The septum of each injection-site cap and the needleless or needle-access cannula were sampled with sterile premoistened swabs. Swabs were cultured on blood agar plates. RESULTS: The rate of fluid pathway contamination was 100% (40/40) for one of the needleless intravenous access devices and 80% (20/25) for the other when septa were contaminated with E. faecium and not disinfected before puncture. The rate for the intravenous needle-access device was 72% (18/25). When the septa of the three different devices tested were disinfected with 70% isopropyl alcohol, E. faecium was isolated on only one septum from all devices tested in part two (1/74, 1.3%). CONCLUSIONS: These laboratory studies demonstrate that there is no statistically significant difference in the rate of fluid pathway contamination between needleless and intravenous needle-access devices. However, if the septa of either needleless or needle systems are not disinfected before puncture, a high rate of fluid pathway contamination may occur.

Ochrobactrum anthropi meningitis in pediatric pericardial allograft transplant recipients.

An epidemiologic investigation was done after 3 patients contracted Ochrobactrum anthropi meningitis at one hospital in October 1994. Neurosurgical patients with pericardial tissue implants were at greater risk of infection than other neurosurgical patients (3/14 vs. 0/566; P<.001). Cultures of implants removed from 2 case-patients, an implant at implantation, a nonimplanted pericardial tissue, and an unwrapped but unopened bottle of Hank's balanced salt solution (HBSS) grew O. anthropi. Patient and tissue isolates had identical genotypes; the isolate from the HBSS bottle had a unique genotype. Culture samples from an unopened HBSS bottle and from pericardial tissue grew Pseudomonas stutzeri of the same genotype; however, no P. stutzeri infections were detected. The investigation documented intrinsic P. stutzeri contamination of HBSS. O. anthropi contamination of tissues occurred during processing, possibly due to extrinsic contamination of HBSS. Active surveillance is needed to detect infection in patients receiving transplanted tissues, and rigorous infection control practice are necessary during tissue harvesting and processing to ensure sterility.

An outbreak of pyrogenic reactions in chronic hemodialysis patients associated with hemodialyzer reuse.

In February 1992, 22 patients undergoing chronic hemodialysis at an outpatient dialysis center experienced pyrogenic reactions (PR). The PR rate was significantly greater (p < 0.001) during the epidemic (February 3-5) than the pre-epidemic period (November 1, 1992-February 1, 1992). All patients with PR used dialyzers that had been manually reprocessed either on February 1 or 3. These dialyzers contained up to 120.8 EU/ml of endotoxin in the blood compartment. The only dialyzer reprocessed before February 1 that was available for analysis was found to contain no detectable endotoxin, while dialyzers reprocessed during the epidemic period contained a median endotoxin concentration of 52.8 EU/ml. The bioburden of water used to prepare dialysate was in excess of the Association for the Advancement of Medical Instrumentation (AAMI) standard for water, < or = 200 colony forming units (CFU)/ml. Samples of treated water collected in the reuse area were within AAMI standards at the time of the investigation (February 11 and February 26), but before the investigation, water samples were assayed with a culture method that could not detect microbial concentrations below 10(3) CFU/ml. In addition, the treated water feed line to the disinfectant container may never have been disinfected. However, no samples were collected from this line during the investigation. This outbreak emphasizes the need to use water that meets the AAMI bacteriologic and endotoxin standards of < or = to 200 CFU/ml and/or 5 EU/ml, respectively, for reprocessing hemodialyzers nad to ensure that appropriate culture techniques are used for treated water dialysate.

Patient cytokine response in transfusion-associated sepsis.

Cytokine concentrations in plasma from patients transfused with packed erythrocytes contaminated with gram-negative bacilli were measured. Cytokine concentrations in posttransfusion plasma were significantly elevated. A difference in cytokine patterns between survivors and a nonsurvivor was observed.

Cytokine kinetics in an in vitro whole blood model following an endotoxin challenge.

Whole blood and peripheral blood mononuclear cell (PBMC) culture models have been used to study cytokine stimulation and release in vitro. In this study, we characterize the kinetics of the interleukins (IL-1 beta), (IL-6), (IL-8), and tumor necrosis factor-alpha (TNF-alpha) following an endotoxin (ET) challenge using our in vitro whole blood model. Whole blood samples from 10 healthy volunteers were studied. All cytokines were measured by enzyme-linked immunosorbent assay. Peak concentrations of TNF-alpha occurred 2 h after ET challenge followed by a rapid decline in free plasma TNF-alpha concentration (half-life 18.2 min). IL-1 beta was not significantly elevated until 4 h after ET challenge. IL-8 was elevated 1 h after ET challenge. IL-6 concentration exhibited a biphasic peak occurring at 6 and 74 h after ET challenge. We conclude that (1) our whole blood in vitro model produces cytokine release kinetics similar to those reported in vivo, and (2) the presence of either binding proteins or cellular metabolism of TNF-alpha in whole blood produces a similar plasma half-life to that observed in vivo.

A prospective study of pyrogenic reactions in hemodialysis patients using bicarbonate dialysis fluids filtered to remove bacteria and endotoxin.

Pyrogenic reactions (PR) are a well-recognized complication of hemodialysis and have been associated with dialyzer reuse, high-flux dialysis, and bicarbonate dialysate. However, the roles of bacteria and endotoxin in dialysate for producing PR are not well defined. To determine the effect of removing most bacteria and endotoxin from the dialysate on the incidence of PR, a cohort of chronic hemodialysis patients receiving high-flux, high-efficiency, or conventional hemodialysis at three centers with bicarbonate dialysis fluids that had been filtered with a polysulfone high-flux hemodialyzer was prospectively studied. Unfiltered bicarbonate concentrate had median bacterial and endotoxin concentrations of 479,000 CFU/mL and 39,800 pg/mL, respectively. After filtration of the bicarbonate concentrate at the central proportioner, dialysate had a median 9.2 CFU/mL of bacteria and 17.8 pg/mL of endotoxin. Dialysate filtered at individual proportioning dialysis machines had a median 0.001 CFU/mL of bacteria and 0.19 pg/mL of endotoxin. Nine PR were identified among 303 patients after 28,007 hemodialysis treatments (0.3 PR/1,000 treatments). The rate of PR was similar for the three hemodialysis treatment modalities and for first-use compared with reused dialyzers. Although the PR rate in this study was lower (P = 0.046) than the PR rate of a previous study with unfiltered dialysis fluids (0.7 PR/1,000 treatments), it represents a difference of only 10 PR in over 28,000 treatments. It was concluded that filtration of hemodialysis fluids is efficacious in removing bacterial and endotoxin contamination and can result in a lower incidence of PR in patients receiving high-flux, high-efficiency, or conventional hemodialysis.

Prestorage removal of Yersinia enterocolitica from red cells with white cell-reduction filters.

Prestorage removal of phagocytic white cells (WBCs) may increase the survivability of contaminating bacteria in units of stored red cells. Fourteen units of whole blood were inoculated with 65 colony-forming units per mL of Yersinia enterocolitica (serotype O:3) and processed into AS-3-preserved red cells. Five red cell units were filtered with a prototype third-generation filter and five red cell units with a second generation filter. WBC reduction was performed on the day of collection. Four red cell units were not filtered. Three noninoculated whole blood units served as negative controls; two were filtered (one with each type of WBC-reduction filter) and one remained unfiltered. All red cell units were then stored at 4 degrees C for 42 days. One of the five filtered red cell units (20%) in each filter group supported growth of Y. enterocolitica. In contrast, 4 (100%) of 4 unfiltered inoculated red cell units had growth (p = 0.04). Overall, 2 (20%) of 10 units of WBC-reduced red cells supported the growth of Y. enterocolitica, as compared to 100 percent of unfiltered red cell units after inoculation (p = 0.015). Bacterial contamination was not detected in any of the three noninoculated units. It can be concluded that prestorage WBC filtration significantly reduces the potential for growth of Y. enterocolitica in red cells stored at 4 degrees C for 42 days.

Pyrogenic reactions in patients receiving conventional, high-efficiency, or high-flux hemodialysis treatments with bicarbonate dialysate containing high concentrations of bacteria and endotoxin.

High-efficiency (HE) and high-flux (HF) hemodialysis are becoming increasingly popular methods for treating patients with chronic renal failure because they reduce the time required for dialysis treatment. HF and HE dialyzers require bicarbonate dialysate, often prepared from concentrates that can support bacterial growth with endotoxin production. There is a concern that endotoxins or bacteria may cross or interact at the membranes of these dialyzers, triggering the release of endogenous pyrogens (cytokines) by peripheral blood mononuclear cells to cause pyrogenic reactions (PR). To determine the incidence of PR and to examine the association between PR and levels of bacteria and endotoxin in dialysate, a cohort of patients receiving conventional, HE, or HF hemodialysis with bicarbonate dialysate and reprocessed dialyzers at three dialysis centers during a 12-month period was studied prospectively. All dialyzers underwent a test of membrane integrity before use. A total of 19 PR were identified among 18 patients in 26,877 hemodialysis treatments (0.7 PR/1,000 treatments). There was no significant difference in PR rates by treatment modality: conventional, 0.5 per 1,000 (7 PR/13,123 treatments) versus HE, 0.9 per 1,000 (9 PR/11,345) versus HF, 1.2 per 1,000 (3 PR/2,409) (P = 0.21; chi 2 test). Throughout the study period, bacterial counts for dialysate at each center significantly exceeded the Association for the Advancement of Medical Instrumentation's (AAMI) microbiologic standards for dialysate of less than 2,000 CFU/mL (mean, 19,000 CFU/mL), but water used in the reuse of dialyzers tested less than 200 CFU/mL.(ABSTRACT TRUNCATED AT 250 WORDS)

Microbial growth and endotoxin production in the intravenous anesthetic propofol.

OBJECTIVE: In this study, we measured microbial growth and endotoxin production in the intravenous anesthetic propofol using 10 different microbial strains; 6 isolated from outbreak cases and 4 from laboratory stock cultures. DESIGN: In each trial, endotoxin-free glass tubes containing 10 ml propofol were inoculated with 10(0)-10(3) CFU/ml of the test organism and incubated at 30 degrees C for 72 hours. SETTING: In May and June 1990, the Centers for Disease Control received reports of 5 outbreaks in 5 states of postsurgical patient infections and/or pyrogenic reactions. Epidemiologic and laboratory investigations implicated extrinsic contamination of an intravenous anesthetic, propofol, as the probable source of these outbreaks. RESULTS: After 24 hours, 9 of the 10 cultures increased in viable counts by 3 to 6 logs. At least 1 ng/ml of endotoxin was produced within 24 hours by Escherichia coli, Enterobacter cloacae, and Acinetobacter calcoaceticus subspecies anitratus. CONCLUSIONS: Propofol can support rapid microbial growth and endotoxin production. To avoid infectious complications, scrupulous aseptic technique should be used when preparing or administering this anesthetic.

Effects of incubation time and temperature on microbiologic sampling procedures for hemodialysis fluids.

To prevent pyrogenic reactions and bacteremia in hemodialysis patients, the Association for the Advancement of Medical Instrumentation and the Centers for Disease Control recommend culturing of hemodialysis fluids (water and dialysate) at least once a month. The recommendations for total microbial counts are (i) less than or equal to 200 CFU/ml in water used to prepare dialysate or reprocess hemodialyzers and (ii) less than or equal to 2,000 CFU/ml for the dialysate. In accordance with the Association for the Advancement of Medical Instrumentation recommendations all cultures should be incubated at 37 degrees C for 48 h on suitable culture media, such as Trypticase soy agar, standard methods agar, or one of several commercially available assay systems. There have been suggestions that lower temperatures and longer incubation might improve the recovery of bacteria from water and dialysate. In this study bacterial recovery from various dialysis fluids (water, bicarbonate dialysate, and bicarbonate concentrate) at 30 and 37 degrees C was compared. Duplicate sets of samples were membrane filtered (pore size, 0.45 microns); one set was incubated at 30 degrees C and the other was incubated at 37 degrees C for 72 h. The number of visible colonies was counted every 24 h by using a dissecting microscope. No significant difference was observed in specimens incubated at 37 degrees C for 48 h compared with those incubated at 30 degrees C for 72 h. Also, bacterial recovery was significantly better when samples of bicarbonate dialysate or bicarbonate concentrate were plated on Trypticase soy agar as opposed to standard methods agar.

Comparison of microbiologic assay methods for hemodialysis fluids.

To help prevent pyrogenic reactions and bacteremia in hemodialysis patients, the Association for the Advancement of Medical Instrumentation and the Centers for Disease Control recommend microbiologic assay of hemodialysis fluids at least monthly. Five commercially available assay systems were evaluated by using the membrane filtration technique with standard methods agar and trypticase soy agar as the standards for comparison. Each assay system was challenged with dialysate and reverse-osmosis water from local dialysis centers, aqueous suspensions of eight laboratory strains of gram-negative bacilli and nontuberculous mycobacteria, and a mixed microbial flora inoculated into reverse-osmosis water and laboratory-prepared dialysate. Mean viable counts from triplicate samples were obtained after incubation at 37 degrees C for up to 72 h. The efficiency of recovery varied with the specific type of microbial challenge. The SPC water sampler (Millipore Corp., Bedford, Mass.) was the most consistent in obtaining the highest viable counts. Other commercial systems were comparable to each other in overall performance. All assay systems tested provided an acceptable balance between microbial recovery and required sampling time, equipment, and expertise.

Mycobacterium chelonae infection among patients receiving high-flux dialysis in a hemodialysis clinic in California.

Between July 1987 and January 1988, five patients dialyzed at a hemodialysis outpatient clinic developed systemic Mycobacterium chelonae abscessus (MCA) infections. Four of the five patients had arteriovenous graft infections, and two died during antimicrobial therapy. Case-patients were more likely than control-patients to have received high-flux dialysis during the 6 mo before their infection (100% vs. 30%, P = .009). MCA was cultured from the hose connected to a water spray device used for manual reprocessing of high-flux but not regular dialyzers. Renalin was the germicide used to manually disinfect dialyzers for reuse and was prepared at a concentration of 2.5%. Laboratory studies documented survival of MCA from two high-flux dialyzers that were reprocessed in a manner similar to that used in the clinic. Early diagnosis with complete rather than partial graft removal in combination with antimicrobial therapy is recommended for MCA graft infections. In addition, 2.5% Renalin did not appear to ensure complete killing of MCA in high-flux dialyzers that were manually reprocessed at this hemodialysis clinic.

Outbreak of gram-negative bacteremia and pyrogenic reactions in a hemodialysis center.

During the period from April 4, 1988, to April 20, 1988, nine pyrogenic reactions and five gram-negative bacteremias occurred in 11 patients undergoing dialysis. All pyrogenic reactions and gram-negative bacteremias occurred among patients in whom a reprocessed dialyzer was used. The rate of pyrogenic reactions or bacteremias per 100 sessions using a reprocessed dialyzer was higher than in sessions during which a new dialyzer was used (4.5 vs. 0; p = 0.03). Dialyzers were manually reprocessed with 2.5% Renalin germicide. The Renalin concentrations varied widely in 12 dialyzers stored after manual reprocessing during the epidemic period (0.9-4.2%); the median endotoxin concentrations varied from 0 to 246 ng/ml and were higher in dialyzers with Renalin concentrations less than or equal to 1.0% than in dialyzers with higher concentrations (p = 0.01). Experiments using a dilution technique described by a technician resulted in Renalin concentrations ranging from 1.4% at the surface to 3.5% at the bottom of the preparation container. These findings suggest that failure to adequately admix Renalin during dilution may be associated with low levels of disinfectant, high levels of bacteria and endotoxins in dialyzers, and outbreaks of pyrogenic reactions and gram-negative bacteremias in dialysis patients.

Recovery of bacteria from reprocessed high flux dialyzers after bacterial contamination of the header spaces and O-rings.

The Centers for Disease Control (CDC) have received reports of bacteremia in patients on high flux dialysis attributed to contamination of dialyzer header spaces or o-rings. A study was performed in which header spaces and o-rings of Hemoflow F-80 dialyzers (Fresenius AG, Bad Homburg, FRG) were exposed to an aqueous suspension of Xanthomonas maltophilia and Mycobacterium chelonae for 1 hour. After exposure, the dialyzers were reprocessed manually with 4% formaldehyde, 4% Renalin, 2.5% Renalin, or sterile water (SW) as a control, or with an automated reprocessing machine using 3.25% Renalin. After 48 hours the blood compartment (BC) was drained and rinsed twice with 500 ml of SW. Each BC sample was cultured. To simulate dialysis, separate circulates of SW were pumped through the DC and the BC. After 15 minutes, the BC circulate was cultured, headers were unscrewed, and o-rings, header caps, and fiber bundle ends were cultured. For each germicide, bacteria were recovered in low numbers, primarily from the o-rings and the o-ring groove in the header caps. In 38 tests, a total of 60 of 342 assays (17.5%) were positive. In only one of these tests one bacterial colony forming unit (cfu) was recovered from the BC circulate during simulated dialysis. It was concluded that if header spaces and o-rings are contaminated, bacteria could be sealed protectively from the germicide. However, concentrations of surviving bacteria were low, probably outside the BC, and did not effectively contaminate the BC circulate during simulated dialysis.

Growth and endotoxin production of Yersinia enterocolitica and Enterobacter agglomerans in packed erythrocytes.

Since 1987, the Centers for Disease Control investigated six cases of transfusion-associated sepsis. All six patients developed septic shock after receiving units of packed erythrocytes (PRBCs) contaminated with Yersinia enterocolitica (five patients) and Enterobacter agglomerans (one patient); three of the blood recipients died. We studied the growth and endotoxin production of Y. enterocolitica and E. agglomerans in units of PRBCs stored at 4 degrees C for 60 days. When PRBCs were inoculated with 0.1 to 1.0 CFU of these organisms per ml, both Y. enterocolitica and E. agglomerans entered log-phase growth 2 to 3 weeks after inoculation; generation times were 15 and 22 h, respectively. Endotoxin was first detected at 3 weeks following inoculation, and the concentration paralleled the log phase of growth of the strains tested. These data show that prolonged storage of PRBCs at 4 degrees C provides conditions that allow these two organisms to grow and subsequently produce high concentrations of endotoxin.

Effect of chemical germicides on the integrity of hemodialyzer membranes.

Epidemiologic investigations of bacteremia in dialysis patients by the Centers for Disease Control (CDC) identified an association with the use of dialyzers disinfected with a specific chemical germicide. A collaborative study by the CDC and the Food and Drug Administration (FDA) was conducted to determine the effect of dialyzer disinfectants on five types of dialyzer membranes: three cellulosic (Cuprophan, cellulose acetate, cuprammonium rayon); and two synthetic (polysulfone, polyacrylonitrile). The disinfectants tested were: 4% formaldehyde; Renalin; Cidex Dialyzer; Sporicidin HO; Warexin; and RenNew-D. Water was the control. Dialyzers were reprocessed up to 15 times. Each reprocessing consisted of rinsing, air-leak testing, filling with fresh disinfectant, and storing for 2 to 4 days. After 15 reprocessings or air-leak failure, each dialyzer was microbiologically challenged for membrane integrity. Membranes exposed to Renalin, Cidex Dialyzer, and water passed all tests. Cellulosic membranes exposed to Warexin failed all tests after 2 to 9 reprocessings. Cellulose acetate membranes exposed to Sporicidin HD failed microbiologic testing. One polysulfone dialyzer exposed to RenNew-D and one exposed to 4% formaldehyde failed microbiologic testing. These results and those obtained from epidemiologic studies suggest that membrane integrity testing (e.g. an air-leak test) should be an integral part of dialyzer reprocessing.