To examine factors associated with nosocomial acquisition of Pseudomonas cepacia in adult patients with cystic fibrosis.

A retrospective case-control study of 5 patients with nosocomial acquisition of P cepacia versus 20 matched controls who failed to develop P cepacia infection. Selective handwashing, air sampling, and respiratory equipment sampling also were performed.

A university hospital providing tertiary care to 95 adult cystic fibrosis patients.

All patients are adults with known cystic fibrosis. Case definition required multiple negative sputum cultures for P cepacia prior to and during admission, with a positive sputum culture prior to discharge. Controls had negative sputum cultures for P cepacia prior to and throughout hospitalization. Controls were matched for age, gender, disease severity, and frequency of hospitalizations.

Factors associated with increased risk of nosocomial acquisition of P cepacia included receiving humidifier or nebulized treatments (60% versus 5%, p = .016, odds ratio= 28.5, 95% confidence interval= 1.93 to 420.58). Factors without significance included ward, room, teaching versus nonteaching status, use of steroids, sharing a hospital room with another cystic fibrosis patient, antibiotic use, presence of portocath in situ, or socializing with another individual with cystic fibrosis known to be P cepacia -positive. Air sampling studies failed to demonstrate aerosolization of P cepacia by coughing cystic fibrosis patients over a 1-hour sampling time. Handwashing studies failed to demonstrate P cepacia on hands of cystic fibrosis patients, nurses, or physiotherapists (before or after physiotherapy). Reservoirs from nebulizers consistently grew P cepacia following therapy.

Respiratory equipment may be an important source of nosocomial acquisition of P cepacia in adult cystic fibrosis patients.

To determine the prevalence of aspergillosis in lymphoma patients housed in a protective environment while undergoing a bone marrow transplant or peripheral stem cell transplant and its relation to lymphoma type, type of transplant, period of neutropenia, method of diagnosis, species of Aspergillus, and the use of empiric amphotericin B.

Clinical, autopsy, and microbiology records were reviewed retrospectively to determine the presence or absence of invasive aspergillosis. All positive specimens underwent further review to determine parameters outlined above.

The review took place at the University of Nebraska Medical Center with lymphoma patients housed in the oncology/hematology special care unit, which consists of 30 single-patient rooms under positive pressure with high-efficiency particulate air filtration.

4 17 lymphoma patients admitted to the oncology/hematology special care unit who underwent 427 courses of high-dose chemotherapy with or without total body irradiation followed by a stem cell rescue.

Twenty-two cases (5.2%) of nosocomial invasive aspergillosis (14 caused by Aspergillus flavus, 2 by Aspergillus terreus, 2 by Aspergillus fumigatus, and 4 by characteristic histology) were diagnosed. The prevalence of disease according to transplant was 8.7% for allogeneic bone marrow transplant (2/23 treatments), 5.6% for autologous peripheral stem cell transplant (9/161), and 4.5% for autologous bone marrow transplant (11/243). Fifteen patients were presumptively diagnosed prior to death (68.2%) most commonly by histologic examination of skin biopsies. All 22 patients received amphotericin B therapy, 17 prior to aspergillosis diagnosis, and 7 (31.8%) survived. No patient with disseminated disease survived.

Even when housing lymphoma patients undergoing myeloablative therapy in a protective environment containing high-efficiency particulate air filtration, there was a risk of developing aspergillosis. These data also showed that antemortem diagnosis with aggressive amphotericin B therapy was most effective in the management of infected lymphoma patients when engraftment occurred and the disease did not become disseminated.

To compare the proportion of recapped needles, an injury surrogate measure, in disposal boxes on two different types of hospital units, both before and after an intervention.

Prospective nonrandomized intervention trial.

A major public teaching hospital.

Specific hospital units. We selected two types of hospital units for study: the first type of hospital unit (medical-surgical ward) had existing mounted in-bathroom needle disposal boxes, and the second type of hospital unit (intensive care unit) had unmounted needle disposal boxes in the room but not necessarily near the patient's bedside.

The installation, in the medical-surgical wards only, of mounted needle disposal boxes on the wall near the patient's bed. The box location in the intensive care units remained the same. In both types of unit, a new mailbox-slot disposal box (SAGE) also was substituted for the previous round-top container.

The baseline proportion of recapped needles in the first medical-surgical unit was 32.6% (±3.8%) and in the second medical-surgical unit it was 27.4% (± 4.0%) in the bathroom needle disposal boxes, which was similar to the observed proportion (34.7 ± 6.4%) in the intensive care unit boxes. Following the intervention, the proportion of recapped needles was significantly reduced in the disposal containers adjacent to the bedside in medical-surgical units, to 27% (a difference of 2.9 standard errors of the baseline distribution) in the first unit and 18.2% (a difference of 4.6 standard errors) in the second. In the intensive care unit, where boxes were not moved but new mailbox-types were simply substituted, no significant change was noted (36.6%, a difference of 0.59 standard errors). A statistically significant reduction was observed in the proportion of needles recapped in both wards combined following the intervention (30.2% to 26.2%, a difference of 2.9 standard errors).

Environmental changes alone are an effective means of altering the risk to healthcare workers from sharp instruments. The use of needle-box counts provides a sensitive and stable instrument to measure injury surrogates and, indirectly, behavioral change in hospital workers.