7588 Background: Although several studies have been reported about non-Hodgkin’s lymphomas (NHL) with bone marrow (BM) involvement, most of them have not performed immunophenotypic studies and contained heterogeneous NHL histologies. Until now, only a few studies with small sample sizes have been reported about clinical characteristics and prognostic factors in diffuse large B-cell lymphoma (DLBCL) with BM involvement. Methods: Between January 1993 and March 2004, 486 patients were diagnosed with DLBCL. Among 84 DLBCL patients who had BM involvement at initial diagnosis, 9 were not ineligible because of the lack of clinical data or unavailability of BM specimen. So, clinical factors and patterns of BM involvement of 75 patients were analyzed in this study. Results: At initial diagnosis, the median age was 57 years (range: 25∼79). In addition to BM, lymph nodes (76%), spleen (23%), Waldeyer’s ring (19%), gastrointestinal tract (16%), lung/pleura (15%), bone (15%), central nervous system (9%), nasal cavity (8%) and liver (7%) were also involved. Among 75 patients, 67 patients received anthracycline-containing chemotherapy; 4 patients received non-anthracycline-containing chemotherapy and 4 could not receive systemic chemotherapy because of combined medical conditions. The median survival was 32.3 months (5-year overall survival [OS]: 38%). In univariate analysis for prognostic factors, high-intermediate or high international prognostic index (IPI), B-symptoms, leucopenia, anemia, thrombocytopenia, pattern of BM involvement (interstitial or diffuse pattern), > 10% replacement of BM area by lymphoma cells, > 10% of large cell infiltration in BM-involved area by lymphoma at initial diagnosis were associated with poor OS (p < 0.05). Multivariate analysis indicated that > 10% replacement of BM area by lymphoma cells (p < 0.001), peripheral thrombocytopenia (p = 0.001) and high-intermediate or high IPI (p = 0.042) were independent predictors of poor OS. Conclusions: To our knowledge, this is the largest study about DLBCL patients with BM involvement. The BM areas involved by lymphoma cells, peripheral thrombocytopenia and IPI at initial diagnosis are independent prognostic factors in these patients. No significant financial relationships to disclose.

The availability of a remote management system, which provides both physiological-related information about the patient and device-related information about the implanted device, would be helpful during in vivo experiments or clinical trials involving artificial heart implantation. In order to be able to monitor the course of the in vivo experiment continuously regardless of the patient's location, an internet-based remote monitoring system was developed, which can monitor physiological-related information such as pressure (AoP, LAP, RAP, PAP) and flow data, as well as device-related information such as current, direction and pump operating conditions. The home care artificial heart monitoring system which we developed consists of four main components, which are the transcutaneous information transmission system (TITS), local monitoring station (LMS), data server station (DSS), and client monitoring station (CMS). The device-related information and physiological-related information can be transmitted in real time from a patient in a remote non-clinical environment to the specialist situated in a clinic depending on the current capabilities and availability of the internet. The local monitoring station situated at the remote site is composed of a data acquisition and preprocessing unit connected to a computer via its RS-232 port, and which communicate using a Java-based client-server architecture. The remote monitoring system so developed was used during an in vivo experiment of the artificial heart implantation for 2 months and performed successfully according to design specifications.

Flow velocity inside the artificial ventricle of a moving actuator type total artificial heart is investigated by the image-processing technique. Optical flow is estimated from two sequential images of flow patterns to find flow-velocity distribution. The region with high velocity inside the artificial ventricle is found from the estimated flow-velocity distribution. Fluid velocity in a simple glass conduit is analysed and can be estimated by calculating optical flow.