Cell block preparation in urine cytology: examination of utility and workflow in an academic practice.

INTRODUCTION: Urine cytology is a common non-invasive test to screen for urothelial carcinoma. Urine cell blocks may sometimes be prepared as a diagnostic aid (eg, to characterize architecture or perform immunohistochemistry). The aim of this study was to determine whether routinely preparing cell blocks on urine specimens improves diagnostic sensitivity. MATERIALS AND METHODS: Three time periods were compared: time period 1 (prior to November 2009; 1437 consecutive selected cases), when cell blocks were rarely prepared; period 2 (November 2009 to May 2010; 1230 selected cases), when cell blocks were prepared on all cases; and period 3 (after May 2010; 1499 consecutive selected cases), when cell blocks were made only when indicated (for samples with substantial cellular pellets or when requested by a pathologist). RESULTS: Patient demographics and the type of specimens received were relatively similar during the 3 time periods. Increased preparation of cell blocks was not accompanied by a notable improvement in specimen adequacy rate, given that <1%, 2%, and 1% of samples were unsatisfactory for the 3 periods. Only the proportion of atypical cases differed during the time periods, being highest in period 1 (23%), but lower in periods 2 and 3. Turnaround time was fastest for period 1 (mean: 47 hours, median: 33 hours), and slower for period 2 and period 3. CONCLUSION: These data show that routinely preparing cell blocks for urine samples did not improve our laboratory's specimen adequacy rate. Nonetheless, cell block preparation on urine samples did help lower the proportion of atypical diagnoses, when routinely or selectively prepared. Because preparation of cell blocks on all urine cases can be costly and only provides minimal added clinical benefit, our recommendation is to rather judiciously utilize cell blocks when screening urine cytology samples.

University of Pittsburgh Medical Center Remains Tracker: A novel application for tracking decedents and improving the autopsy workflow.

All hospitals deal with patient deaths. Multiple departments and personnel must be coordinated to ensure that decedents are safely managed. Prior to 2004, at the University of Pittsburgh Medical Center (UPMC), when a patient passed away, the process of alerting involved personnel, transporting the decedent, and tracking the completion of clinical documents was cumbersome and inefficient. In order to address these concerns, UPMC Remains Tracker, a web-based application, was developed to improve the efficiency and simplify the logistics related to the management of patient deaths. The UPMC Information Services division developed UPMC Remains Tracker, an application that tracks decedents' locations, documentation status, and autopsy status within UPMC hospitals. We assessed qualitative improvement in decedent remains tracking, decedent paperwork management, and staff satisfaction and compliance. UPMC Remains Tracker improved the process of tracking decedents' locations, identifying involved personnel, monitoring autopsy requests, and determining the availability for funeral home transportation. Resident satisfaction with UPMC Remains Tracker was generally positive and scored as "Improved efficiency" and makes identifying and tracking decedents "Much easier". Additionally, the nursing staff reacted favorably to the application. A retrospective review of the use of the application in the management of 100 decedents demonstrated a 93% compliance rate. Among the cases requiring an autopsy, there was a 90% compliance rate. The process of tracking decedents, their paperwork, involved staff, and decedent autopsy status is often inefficient. This assessment suggests that incorporating new technologies such as UPMC Remains Tracker into the management of hospital deaths provides accurate tracking of remains, streamlines the administrative tasks associated with deaths, and increases nursing and resident satisfaction and compliance.

Experience with voice recognition in surgical pathology at a large academic multi-institutional center.

There are few reports of institutional use of voice recognition technology in clinical practice. We describe our experience with voice recognition-integrated synoptic-like dictation, associating templates with key spoken phrases, that we have used in gross examination of common specimens and as a major component of our workflow since 2001. The primary application is VoiceOver Enterprise (Voicebrook, Lake Success, NY), which uses Dragon NaturallySpeaking Medical Edition (Nuance Communications, Burlington, MA) as its speech engine. This integrates with the anatomic pathology laboratory information system (APLIS) and other applications, such as Microsoft Office (Microsoft, Redmond, WA). The largest user group, pathology assistants, mainly dictates biopsy reports, numbering approximately 210,000 specimens since 2001. The technology has been useful in our anatomic pathology workflow and provided a good return on investment, including marked improvements in turnaround time, results standardization, error reduction, and cost savings. The most helpful features of the software are templating, the seamless integration with APLIS, and the voice command creation tools.