Detection of Bjork-Shiley convexo-concave heart valve outlet strut single leg separations: Consensus image acquisition and interpretation using two different cineradiographic imaging technologies

Background: Conventional cineradiology has been used clinically to detect p artially broken outlet struts (single leg separations, SLSs) in normally fu nctioning Bjark-Shiley convexo-concave (C/C) heart valves. The value of rad iographic screening has, however, been debated in the medical literature. T his study uses the same radiographic technique in sheep implanted with know n-status C/C valves in combination with a newly developed geometric image m agnification radiography system. This study was designed to test whether se nsitivity and specificity of radiographic screening of C/C valves in detect ing SLSs could be improved through the combination of readers and imaging m odalities. Methods: Twenty-one sheep with mitral C/C valves were studied on both systems. Five were used for extensive scanning training. When operato rs were expert on both systems, 16 blinded study valves (4 intact and 12 wi th outlet strut SLSs) were scanned twice on both systems, first on a modifi ed conventional and then a prototype geometric image magnification (Feinfoc us(TM)) cineradiographic system by two expert physicians working together. Results: Among the 32 scanned valves, the two combined expert physicians we re required to evaluate 40 intact legs and 24 with an SLS. For all SLS valv es, the conventional and Feinfocus systems separately detected 50 (12/24) a nd 54% (13/24), respectively. When the two systems were combined, the final consensus score was correct in 67% (16/24) of a II SLS valves. Conclusions : Combined modality, paired expert physicians detected 67% of all SLSs. The Feinfocus system might be best reserved for those patients in whom the Sie mens screening study demonstrates in minimally suspicious (grade 2) or susp icious (grade 3) appearance of a C/C valve outlet strut leg.