To estimate the risk of exposure and infection with bloodborne pathogens, a seroepidemiologic survey was conducted among funeral service practitioners (FSPs) in Maryland.

Of 262 members of the Maryland State Funeral Directors Association, 130 (49%) volunteered to participate in the study. In addition to a brief questionnaire, designed to assess both occupational and non-occupational risk factors for bloodborne pathogen infection, participants were screened for markers of human immunodeficiency virus (HIV), hepatitis C virus (HCV), and past hepatitis B virus (HBV). Titers for antibodies to hepatitis B surface antigen (anti-HBs) also were examined and compared with history of hepatitis B vaccination.

Seroprevalence for HIV, HBV, and HCV infection was 0.8%, 4.6%, and 0%, respectively. Nearly 19% of participants reported at least one bloodborne exposure in the past 6 months. The one HIV infection and all but two of the HBV infections were correlated with well-established non-occupational risk behaviors. Disposable gloves were worn by 96%, and eating, drinking, or smoking during embalming were infrequent. Sixty-one percent of FSPs reported having received one or more doses of hepatitis B vaccine at some time in the past. Of those who reported having received all three doses of vaccine, 67% had adequate titers to hepatitis B surface antibody, the marker of protection related to vaccination.

Compared with prior studies of FSPs, this study found a low rate of occupational exposures and a high rate of hepatitis B vaccination, suggesting improved compliance with recommendations for preventing transmission of bloodborne pathogens in the workplace.

To investigate an outbreak of surgical site infections (SSI) in a vascular surgery unit.

A 60-bed unit of vascular surgery, where surgeons performed an average of 30 operations per month at the Hospital do Servidor Público Estadual, a 1,000-bed tertiary care hospital in São Paulo, Brazil.

We included in the case group nine patients who had limb amputations or arterial reconstructions, October 16 through 23, 1992. We included in the control group patients whose operations were performed within 30 days of the outbreak period. Control patients were matched for sex and type of operation.

Six of 9 case patients experienced SSI, as compared with 3 of 18 control patients (P= .026) and 28 of 244 patients in the pre-epidemic period (P= .0002). Risk factors were identical for case and control groups. Factors assessed were American Society of Anesthesiology (ASA) status, duration of surgery, wound class, emergency status, remote site infections, preoperative length of stay, use of prophylactic antibiotics, and underlying diseases. Possible common sources also were analyzed. No differences were observed concerning hair removal, preoperative shower, wound dressing, and surgical team present in the operating room. During the outbreak period, the operating room was not provided with povidone-iodine, used in our hospital for skin cleansing and handscrubbing. Surgeons from all departments, including vascular surgery, used 2% iodine with 70% alcohol for skin cleansing. Surgeons from other departments used this iodine solution for handscrubbing, but the vascular surgeons used plain soap for handscrubbing. No increases in SSI rates were reported in other services. Comparison of case and control groups for handscrubbing was statistically significant (P<.00001). After reinstitution of povidone-iodine, only one SSI was diagnosed in 13 vascular procedures.

Although we could not demonstrate definitely that scrubbing with plain soap was related to SSI, we found a strong suggestion of this association

To validate a previously published model for predicting bacteremia in hospitalized patients.

Application of a published bacteremia prediction model to a prospective validation cohort of patients and comparison of its predictability to that found in the derivation cohort.

Urban, university-affiliated, 550-bed public hospital.

The validation cohort consisted of 342 patients with 559 blood culture episodes between October 14, 1992, and December 5, 1992. Each blood culture episode was scored based on the presence or absence of seven predictors of bacteremia and the findings compared with published results (derivation cohort).

None.

Application of the bacteremia prediction model to the validation cohort identified episodes with a low risk (3%) and a high risk (17%) for true bacteremia, similar to the findings in the derivation cohort (1% and 16%, respectively). Comparison of the predictions of the model in the two cohorts by receiver operator characteristic curve analysis revealed that the overall predictability of the model in the validation cohort was not as good as in the derivation cohort.

Although the bacteremia prediction model did not perform as well overall in the validation cohort, the model still was able to clearly define two extreme groups: those with a low risk and those with a high risk for true bacteremia. This predictive capability may aid physicians in prescribing empiric antimicrobial therapy and also may be useful to hospital epidemiologists in assessing quality of care

To determine the prevalence, incidence, and etiology of conjunctivitis in residents of a long-term care facility.

Prospective surveillance of episodes of conjunctivitis during two 6-month periods.

Three-building, 319-bed, multipurpose, long-term care facility including units for personal care, chronic care, palliative care, rehabilitation, respiratory rehabilitation, and chronic ventilator patients.

The incidence of conjunctivitis on different units varied from 0.6 to 3.5 per 1,000 patient-days. One building had a significantly higher rate of acute conjunctivitis and, within that facility, the most highly impaired patients had significantly more disease. Patients with chronic conjunctivitis generally were admitted with conjunctivitis; institutionally acquired conjunctivitis proceeding to chronic conjunctivitis seldom was observed except on the chronic ventilator unit. Residents with chronic conjunctivitis were significantly more likely to have a diagnosis of glaucoma, ectropion, or entropion. Potential bacterial pathogens were isolated from 3 (21%) of 14 and 24 (38%) of 69 acute episodes for which cultures were available in the two study periods. Staphylococcus aureus was the most frequent pathogen isolated. Empiric therapy was usually with topical sodium sulamyd or gentamicin, with substantial interphysician variability in prescribing patterns. Symptom duration did not differ for episodes which did or did not receive topical antimicrobials.

Conjunctivitis is common in this facility and occurs with increased frequency in more highly impaired residents. Most episodes may not be due to bacterial infection. Further study of optimal management approaches to conjunctivitis is necessary

To describe the frequency and patterns of infection caused by methicillin-susceptible Staphylococcus aureus (MSSA) and methicillin-resistant S aureus (MRSA) infections in a single nursing home population and to determine the effect of MRSA's entry into the facility on subsequent experience with both MSSA and MRSA infections.

Observational and descriptive. Surveillance data on nursing home-acquired infections were reviewed to identify all patients with MSSA and MRSA infections occurring during the 5-year period from 1987 to 1991. The medical records of these patients were reviewed retrospectively to collect additional information about the patients and their infections.

A 120-bed Veterans' Affairs nursing home care unit (NHCU) whose residents predominantly were elderly men with severe underlying diseases and functional impairments,

During the 5-year study period, 40 MSSA and 28 MRSA infections were acquired by NHCU residents. Twelve to 19 S aureus infections occurred each year. S aureus accounted for 13% to 17% of all NHCU-acquired infections during the years of study, occurring with a frequency of 0.29 to 0.47 infections per 1,000 resident-care days. MRSA infections, first detected in 1988, accounted for an increasing percentage of S aureus infections in subsequent years, but this increase had little effect on the facility's overall infection rates, the composite S aureus infection rates, or the types of infections observed. MSSA and MRSA infections acquired in the NHCU were comparable. Both affected patients with severe underlying diseases and functional impairments. Pneumonia, urinary tract infections, skin and soft tissue infections, and conjunctivitis were the types of infections observed most frequently, accounting for 28%, 25%, 22%, and 15% of all S aureus infections, respectively. Four bacteremic infections occurred in the MSSA group, and five in the MRSA group (P=.47). Four of the MSSA and three of the MRSA infections resulted in death (P= 1.0). Nine of the MSSA and 12 of the MRSA infections resulted in the patient's transfer to the associated acute care hospital for additional care (P= .13).

In the NHCU setting, MSSA and MRSA infections were similar in terms of the types of residents affected, the sites involved, and the frequency of adverse outcomes. The entry of MRSA into the facility appeared to have no effect on the subsequent experience with NHCU-acquired infections caused by S aureus.

To study the spread of strains of Enterobacter aerogenes in our hospital in 1992 and 1993 by using two genotypic markers, and to evaluate these methods for the epidemiological investigation of this species.

Ribotyping (using two endonucleases) and arbitrarily primed (AP)-PCR (using two different 10-mer primers) were applied to the epidemiological typing of clinical strains of E aerogenes isolated from hospitalized patients.

The intensive care unit (ICU; 5 patients, 13 isolates), nephrology units (3 patients, 5 isolates), and surgery units (2 patients, 2 isolates) of the university hospital of Saint-Etienne (France).

Eight epidemiologically unrelated isolates, chosen as controls, exhibited distinct profiles, both by AP-PCR and ribotyping. Two clones of E aerogenes circulated in the ICU; both were isolated successively from samples of a single patient who stayed in the unit for almost 1 year. A third clone was recovered from patients of surgery units. A fourth clone was shown to have infected patients of nephrology units.

Ribotyping and AP-PCR appear to be reliable methods for typing E aerogenes strains implicated in nosocomial infection. The spread of independent clones of E aerogenes in different units of our hospital in 1992 and 1993 was demonstrated by both methods. This study emphasizes the need to choose the endonucleases or primers with care to obtain high discriminatory results in genotypic investigations.