Liquid-based rapid onsite evaluation of endobronchial ultrasound cytologies.

INTRODUCTION: Rapid onsite evaluation (ROSE) generally uses smears made at the site of the procedure ("smear-based ROSE"). It requires considerable time, generally 2 individuals, technical expertise, and it can be difficult to estimate material available for ancillary studies. We developed an alternative ROSE using liquid-based cytology ThinPrep with hematoxylin and eosin (H&E) stain ("liquid-based ROSE") and assessed its advantages. MATERIALS AND METHODS: Clinicians rinse the sample(s) into CytoRich Red and send to Pathology. A defined proportion of the needle rinse is removed for a ThinPrep stained with a rapid H&E. Adequacy and diagnosis were compared to final outcome. Total time was recorded. RESULTS: Among 52 liquid-based ROSE readings, 28 (53.8%) were interpreted as "adequate" with final as adequate; 17 (32.7%) were interpreted as "inadequate" with final as inadequate; 7 (13.5%) were interpreted as "inadequate" with final as adequate. Of 23 readings provided with onsite diagnosis, 15 (65.2%) were interpreted as definitive positive or negative diagnoses; 6 (26%) were interpreted as nondiagnostic; and 2 (8.7%) were interpreted as atypical. All definitive diagnoses were concordant with final diagnoses. The time for liquid ROSE performance ranges from 6 to 22 minutes (mean: 13 minutes) and required only 1 individual. CONCLUSIONS: Liquid-based ROSE allows accurate adequacy determination and diagnosis, takes about 15 minutes of cytologist time, and can be performed by just 1 person. The technique produces well-preserved and stained slides, it may allow a better estimation of the total amount of material in the specimen vial and may provide a better platform for telecytology.

Training family practice residents in HIV care.

Although family physicians are likely to encounter patients at risk for or diagnosed with HIV, they rarely are trained in prevention and treatment techniques. From 1997 to 2000, the University of Minnesota developed and evaluated a longitudinal curriculum to train family practice residents in the prevention and comanagement of HIV disease. Five residency programs participated. The core curriculum was based on 18 teaching modules incorporated into the standard 3-year residency program and covers topics in prevention, diagnosis, and treatment of HIV disease. Residents were encouraged to participate in elective rotations of HIV clinical care, service, and research, as well as attend a Sexual Attitude Reassessment Seminar. Baseline and postcurriculum attitude and knowledge surveys were performed and scores were compared before and after completion of the curriculum, as well as between the participants and a historical control group of residents who graduated in 1997. During the 3 academic years from 1997 to 2000, 214 residents participated in the study. Entering classes did not differ in baseline knowledge. Residents' knowledge about HIV disease, attitudes, confidence, and intention to treat HIV-positive patients improved (p < 0.01). The yearly number of HIV enzyme-linked immunosorbent assay (ELISA) tests performed at the residency sites increased from 1145 to 1665 by the project's end. The HIV Curriculum Project provides a useful model of an interdisciplinary program for training family medicine residents in HIV/AIDS. The evaluation data provide an assessment of the curriculum itself and an important look at family practice residents' knowledge, attitudes, and performance related to HIV over time.