The epidemiology of infection with Pseudomonas pyocyanea in a burns unit

The epidemiology of burn wound infections: then and now

Burn wound infections are a serious complication of thermal injury. Although pneumonia is now the most important infection in patients with burns, burn wound infection remains a serious complication unique to the burn recipient. The methods for managing thermal injury have evolved during the past 50 years. This evolution has been accompanied by changes in the etiology, epidemiology, and approach to prevention of burn wound infections. In the 1950s, 1960s, and 1970s and into the mid-1980s, burn wounds were treated by the exposure method, with application of topical antimicrobials to the burn wound surface and gradual debridement with immersion hydrotherapy. As early burn wound excision and wound closure became the focal point of burn wound management, accompanied by a change from immersion hydrotherapy to showering hydrotherapy, the rate of burn wound infection appeared to decrease. Few epidemiologic studies have been done since this change in the approach to management of thermal injury. There are few data on the epidemiology of burn wound infections from the era of early excision and closure. Data are needed on infection rates for excised and closed burn wounds, the etiologies of these infections, and the epidemiology and the prevention of such infections. Additional studies are needed on the indications for topical and antimicrobial prophylaxis and selective decontamination of the digestive tract.

PYOCINE-TYPING OF HOSPITAL STRAINS OF PSEUDOMONAS PYOCYANEA

A method of typing Pseudomonas pyocyanea is described which makes use of the property of bacteriocine production. The method was applied in an epidemiological study of Ps. pyocyanea infection in a general hospital. It was found that the majority of strains believed on epidemiological grounds to be of common origin were allotted to the same pyocine type. In the course of this study, Ps. pyocyanea was isolated from 219 patients; 474 strains from 149 of these were typed. There was evidence of cross-infection among patients in a urogenital surgical ward, but in other parts of the hospital there appeared to be little cross-infection. A number of individual instances of auto-infection, possibly arising from the bowel flora, were seen.

Outbreak of infection in a burns unit due to Pseudomonas aeruginosa originating from contaminated tubing used for irrigation of patients

Five patients with extensive deep burns developed septicaemia due to Pseudomonas aeruginosa serogroup O-7.8 and phage type 21 or 21/188 shortly after they had been admitted to hospital. Four other burned patients became colonized with the same strain. The source of infection was contaminated tap water used for irrigation of the burns, as part of the first-aid treatment which the patients received when entering the hospital. Contamination was restricted to showers and tubing that were permanently connected to the taps, and the outbreak stopped after they had been disinfected. Tubing and showers used for irrigation of burns should be dismantled and heat-disinfected after each patient and not reconnected to the taps until immediately before the next treatment. Taps used for irrigation of burns should be monitored regularly for the presence of P. aeruginosa and other potentially pathogenic bacteria. Routine typing of P. aeruginosa isolates from burned patients is indicated in order to detect and eliminate hidden sources of infection.

Oropharyngeal and fecal carriage of Pseudomonas aeruginosa in hospital patients

This prospective study was designed to determine the incidence of rectal and/or oropharyngeal colonization rates of patients with Pseudomonas aeruginosa upon admission to a general hospital and the risk of becoming colonized while hospitalized. Consecutive 186 admissions (180 patients) to one medical ward, one surgical ward, and the intensive care unit were studied over a period of 5 months. Rectal and oropharyngeal swabs for P. aeruginosa were obtained on admission, weekly thereafter, and/or upon discharge. Forty-two patients (22.6%) were colonized on admission, 20 patients (10.8%) acquired P. aeruginosa during hospitalization. Colonization on admission was observed twice as frequently on the surgical ward and in the intensive care unit as on the medical ward. Positive rectal cultures were more frequent than oropharyngeal cultures throughout the study (P less than 0.01). For patients admitted culture positive or culture negative, the probabilities of remaining culture positive or culture negative, respectively, remained at 44 and 72% after 35 days of hospitalization. The most common P. aeruginosa serotypes were 1, 6, and 10, and pyocin types 1, 3, and 10 were predominant. There was no statistical difference in the serotypes or pyocin types detected on admission or acquired during hospitalization. Except for two hospital-acquired first isolates which were resistant to moxalactam, all first isolates were susceptible to the four antibiotics tested. During the study, one isolate became resistant to azlocillin, gentamicin, and tobramycin, while two isolates became resistant to moxalactam. A statistical analysis was performed for 13 risk factors for all colonized and noncolonized patients. Colonization detected at the time admission was positively associated with age ( > 65 years), previous surgery of the gastrointestinal tract for neoplasm, and anemia ( P< 0.05). In contrast, for patients who entered the study culture negative, none of the analyzed 13 risk factors was associated with an increased probability for colonization. This observation included the administration of antimicrobial agents singly or in combination or both.

Infections in burned patients

Prevention and treatment of burn wound infection requires knowledge of the epidemiology of such infections. Prevention of infection rests on removal of reservoirs or sources of microorganisms from the burn patient's environment and interruption of transfer of microorganisms to the surface of the wound. When prevention fails and burn wound infection develops, successful therapy may depend on an understanding of the epidemiology of the burn wound during therapy. Contrary to the oft stated concept that antibiotics penetrate the avascular burn wound poorly, our study demonstrated that gentamicin and tobramycin achieved therapeutic concentrations in burn wound tissue. As in other types of infections, susceptible microorganisms were eradicated and resistant microorganisms persisted. Of most importance was the observation that resistant microorganisms may repopulate the wound within four days of starting therapy. It would appear that failure of therapy is not due to failure of antibiotics to penetrate the burn wound but rather to rapid development of superinfection during therapy.

Epidemiology of Pseudomonas aeruginosa infection and the role of contamination of the environment in a cystic fibrosis clinic

In order to identify the possible reservoirs and routes of cross-infection with Pseudomonas aeruginosa, samples from patients, staff, and the environment of a cystic fibrosis centre and two control wards at an infectious disease clinic were collected during a two-week period in 1980. All the Ps. aeruginosa strains were phage and serotyped. Ps. aeruginosa was isolated from 90 (51%) of the cystic fibrosis patients and most belonged to the 0-3/9 complex, characteristic of strains from patients in the centre. Some of the patients were able to spread Ps. aeruginosa into the air and to their hands by coughing, and Ps. aeruginosa in dried sputum could survive for at least one week. Strains of the same epidemiological types as found in the cystic fibrosis patients were isolated from sinks, soap, baths, toys, tables, brushes, cloths, and air in the clinic. In contrast, Ps. aeruginosa of the same epidemiological types were only found in a few of the sinks in one of the control wards where a few cystic fibrosis patients were regularly treated in isolation cubicles. The precautions employed to prevent future cross-infection include segregation of Ps. aeruginosa-infected from non-infected patients in separate wards and arranging for visits on separate days in the out-patients clinic. The survival of cystic fibrosis patients treated in the centre is much longer than those treated outside the centre despite the problems of cross-infection.

Hand carriage of aerobic Gram-negative rods by health care personnel

A quantitative culture technique (hand washed in a glove containing broth for 30 s) was used to determine the frequency of hand carriage of aerobic Gram-negative rods by various groups of health care workers and 104 control subjects. Overall, 31% of health care workers carried aerobic Gram-negative rods on their hands compared to 59% of control subjects (P < 0.001). Enterobacter agglomerans accounted for 40% of the isolates, and other Enterobacter spp. 7%. Other organisms included Acinetobacter calcoaceticus 21%, Serratia spp. 11%, Klebsiella spp. 10%, Moraxella spp. 3%, Pseudomonas spp. 3%, Proteus spp. 1.5%, Escherichia coli 1%; Morganella morganii, Citrobacter freundii, Aeromonas sp. and an isolate that was not speciated accounted for 0.5% each. We conclude that endemic hand carriage of aerobic Gram-negative rods by health care personnel is common, but significantly less than that of control subjects. Enterobacter agglomerans is found so frequently on the hands of control subjects that it must be considered part of the normal hand flora.

Microbial contamination of topical medicaments used in the treatment and prevention of pressure sores

Topical medicaments used in the treatment and prevention of pressure sores in patients in three hospitals were examined for Pseudomonas aeruginosa and Staphylococcus aureus contamination. Contamination rates were found to vary between hospitals and were affected by differences in the packaging of the product and in the method of application used by the nursing staff.

Introduction of Pseudomonas aeruginosa into a hospital via vegetables

Pseudomonas aeruginosa was isolated from tomatoes, radishes, celery, carrots, endive, cabbage, cucumbers, onions, and lettuce obtained from the kitchen of a general hospital, with tomatoes yielding both highest frequencies of isolation and highest counts. Presence of P. aeruginosa on the hands of kitchen personnel and cutting boards and knives which they used suggests acquisition of the organism through contact with these vegetables. It is estimated that a patient consuming an average portion of tomato salad might ingest as many as 5 x 10(3) colony-forming units of P. aeruginosa. Pyocine types of P. aeruginosa isolated from clinical specimens were frequently identical to those recovered from vegetables, thus implicating tomatoes and other vegetables as an important source and vehicle by which P. aeruginosa colonizes the intestinal tract of patients.

Epidemiology of Pseudomonas aeruginosa in a burns hospital: surveillance by a combined typing system

For 3 months, 259 cultures of Pseudomonas aeruginosa isolated from nonpatient environmental sources and 262 cultures from 16 infected patients in the Intensive Care Unit (ICU) of Shriners Burns Hospital were typed by a combined system with a high degree of reliability. Sinks were major sources of environmental contamination. Serotypes 1 and 2 were the predominant types found in patients, and they were most prevalent among typable strains from sinks. Strain designations were made on the basis of similarities in data from serological and phage typing. All nontypable strains were typed by pyocin production. Two infected patients carried different strains of P. aeruginosa that remained the same type for 45 days, even though their beds in ICU were approximately 6 feet apart. Cross-contamination from patient to patient and spread of infection by nursing personnel were eliminated as major modes of transmission because nasopharyngeal swabs, hair samples, and hands of nursing staff were consistently negative. Splashing of water from contaminated sinks to fomites was suggested as a possible mode of transfer for this infectious agent.

Mode of transmission of Pseudomonas aeruginosa in a burn unit and an intensive care unit in a general hospital

The transmission of Pseudomonas aeruginosa was studied in the burn unit and the intensive care unit of a 650-bed hospital. There was a tendency among patients in the burn unit to yield more than one type of P. aeruginosa, and several patients shared the same types at a particular point in time, suggesting cross-contamination among patients. Similar observations were made in the intensive care unit. Cultures from the hands of nurses caring for these patients yielded the same types of P. aeruginosa, suggesting the direct handling of patients by the nursing personnel to be the principal mode of transmission.

Protective isolation in a burns unit: the use of plastic isolators and air curtains

The use of plastic isolators and of an ;air curtain' isolator for protection of patients against infection was studied in a burns unit.Preliminary bacteriological tests showed that very few airborne bacteria gained access to a plastic ventilated isolator; even when the filter and pre-filter were removed from the air inflow, settle-plate counts inside the isolator were much lower than those in the open ward, but the difference was smaller in tests made with an Anderson air sampler, which showed also that fewer large bacteria-carrying particles appeared inside the isolator than outside it. An open-topped isolator allowed virtually free access of bacteria from ambient air. The numbers of airborne bacteria inside an air curtain were appreciably lower than the counts of airborne bacteria in the open ward, but not as low as those in the plastic ventilated isolator.Controlled trials of isolators were made on patients with fresh burns of 4-30% of the body surface; the patients were given no topical chemoprophylaxis against Staphylococcus aureus or Gram-negative bacilli. Patients treated in plastic isolators showed a significantly lower incidence of infection with Pseudomonas aeruginosa than those treated in the open ward; this protective effect was shown by isolators with or without filters or with an open top. Ventilated isolators, which protected patients against personal contact and airborne infection, gave a limited protection against multi-resistant ;hospital' strains of Staph. aureus, but no such protection was given by an open-topped isolator, which protected only against personal contact infection, or by air curtains, which protected only against airborne infection; the air curtain gave no protection against Ps. aeruginosa, and there was no evidence of protection by any isolator against Proteus spp. and coliform bacilli.Both the controlled trials and evidence from the bacteriology of air, hands, fomites and rectal and nasal swabs taken on admission and later, supported the view that Ps. aeruginosa is transferred mainly by personal contact, Staph. aureus probably by air as well as by contact and coliform bacilli mainly by self infection with faecal flora, many of which are first acquired from the hospital environment in food or on fomites.The use of plastic isolators is cumbersome, and of limited value except in the control of infection with Ps. aeruginosa. For this reason and because of the effectiveness of topical chemoprophylaxis such isolators are unlikely to have more than an occasional use in the treatment of burns. Though air curtains greatly reduce airborne contamination, their use in a burns unit does not appear to protect patients against infection when the alternative (and, for Ps. aeruginosa, more important) routes of contamination by personal contact and fomites are left open.

Microbiological evaluation of protected environments during patient occupancy

Microbiological monitoring has been conducted in two life island (LI) units and two laminar airflow (LAF) rooms while they were occupied by patients undergoing cancer chemotherapy. There were only 5 organisms per 1,000 ft(3) of air sampled in LAF rooms, 31 organisms in LI units, and over 3,000 organisms in regular hospital rooms. None of the floor samples obtained from hospital rooms was sterile, compared to over 70% in LAF rooms. The rate of deposition of organisms onto settling plates was one organism per 4.5 hr in LAF rooms compared to one organism per 0.08 hr in hospital rooms. Potential pathogens were isolated much more frequently from environmental samples obtained from hospital rooms than from LI units or LAF rooms. Two sites of persistent contamination arose in the LAF rooms: the vinyl tile flooring and the water supply system. Over half of the potential pathogens cultured from the protected environment units were cultured initially from the patients who occupied the units.

The effect of a changed environment on bacterial colonization rates in an established burns centre

In an established burns centre which moved from an old building to new purpose-designed premises, colonization rates of patients' burns with Staphylococcus aureus, Pseudomonas aeruginosa and other Gram-negative bacilli were not reduced. Colonization rates with Streptococcus pyogenes increased but the increase was mainly due to multiple importations in the new premises of a strain of higher communicability than any seen in the old.In the first 32 months in the new environment 10 patients were found colonized with pseudomonas on admission and 20 became colonized in the unit. A much higher proportion of patients with burns of more than 30% body surface became colonized than of patients with less. About one-third of the above 20 patients became colonized with strains already isolated from another patient; all but one of them had small area burns. Cross-infection was not observed from numerous heavily colonized patients with high percentage burns. This paradox is discussed in detail. Basin outflows in the new premises became colonized with P. aeruginosa of two serotypes not found on patients in this unit.

Gram-negative bacilli in burns

In a period of two years, 865 strains of Gram-negative bacilli other than Pseudomonas aeruginosa isolated from burns were identified by a range of tests. The commonest species were Proteus mirabilis, Escherichia coli, and Enterobacter cloacae. Many strains of Klebsiella aerogenes, Enterobacter aerogenes, and Bacterium anitratum were also found.A large proportion of the strains were tested for sensitivity to nalidixic acid, ampicillin, kanamycin, chloramphenicol, tetracycline, and carbenicillin, and smaller numbers of strains were tested for sensitivity to cephaloridine, polymyxin, streptomycin, sulphadiazine, sulfamylon, and trimethoprim. The proportion of strains sensitive and resistant to different antibacterial agents varied widely with species of bacteria. A large proportion of the strains of E. coli and P. mirabilis were resistant to ampicillin, which was much used in treatment; resistance appeared least often towards nalidixic acid, kanamycin, trimethoprim, and gentamicin. Multiple resistance occurred less often among strains of E. coli than among Klebsiella spp, Enterobacter spp, and P. mirabilis. Phage and serological typing of Ps. aeruginosa showed that most infections of burns with this organism were due to strains previously found in other patients in the same ward. Taken with other evidence, this supported the view that most infections with Ps. aeruginosa were not acquired from the patient's own flora but from sources in the hospital environment.

Persistence of Salmonella typhimurium on fabrics

The persistence of Salmonella typhimurium (V-31) on wool blanket, wool gabardine, cotton sheeting, cotton knit jersey, cotton terry cloth, and cotton wash-and-wear fabrics was studied. Three methods of exposure were employed to contaminate the fabrics: direct contact, aerosol, and a lyophilized mixture of bacteria and dust having a high content of textile fibers. After contamination, the fabrics were held in 35 or 78% relative humidity at 25 C. The persistence time of S. typhimurium on fabrics held in 35% relative humidity was substantially longer when the fabrics were contaminated by direct contact or by exposure to dust containing bacteria than when contaminated by exposure to aerosolized cultures. Viable bacterial populations persisted for 24 weeks at relatively high population densities on swatches of wool gabardine, cotton sheeting, cotton knit jersey, and cotton terry cloth exposed by direct contact and held in a humidity of 35%. In 78% humidity, bacterial populations persisted on the fabrics for relatively shorter periods of time regardless of the mode of contamination or fabric type. This organism retained its virulence for Swiss mice after being recovered from wool gabardine swatches held 8 weeks in humidities of 35 or 78% and from cotton terry cloth swatches held 6 weeks in the same humidities.

Survival of wound pathogens under different environmental conditions

Suspensions of Pseudomonas aeruginosa, Staphylococcus aureus, Escherichia coli, Klebsiella aerogenes and micrococci were tested for survival on drying and after drying in the atmosphere. The proportion of Gram-negative bacilli that died during drying was greater than that of the Gram-positive cocci, but survival after drying was similar in the two groups of organisms. There were considerable differences in the death-rates on drying of different strains of bacteria, which were consistent on replicate sampling.

A suspension of Ps. aeruginosa prepared from cells which had survived one drying showed a considerably higher proportion of survivors on a second drying; suspensions prepared from subcultures of the dried cells, however, showed a death-rate on drying which was similar to that of the original culture.

Strains of Ps. aeruginosa isolated from floor dust showed a significantly higher proportion of survivors immediately after drying than strains from patients, but the proportion of survivors after 24 hr. of exposure to the atmosphere was approximately the same in the two groups.

Tests were made for survival of Ps. aeruginosa and Staph. aureus in deionized water, tap water, physiological saline and Ringer's solution. In deionized water, Ps. aeruginosa showed a rapid initial loss but some survival for several weeks; Staph. aureus, on the other hand, showed an initial increase in numbers but no survivors after 48 hr. In Ringer's solution all strains of Ps. aeruginosa multiplied rapidly and survived for many weeks; Staph. aureus died rapidly and no strain could be detected after 4 days.

We are grateful to Dr M. T. Parker and Mrs E. Asheshov for the typing of strains of Pseudomonas aeruginosa.