During chordal-sparing mitral valve replacement (MVR), some recommend anatomic reattachment of the anterior leaflet chordae to the anterior annulus; others advocate shifting the chordae to the posterior annulus. To compare the results of these techniques with those of conventional MVR (total chordal excision), 21 dogs were studied 5 to 12 days after implantation of tantalum markers to measure left ventricular volume and geometry. One to 3 weeks later, animals underwent conventional MVR (n = 7) or chordal-sparing MVR with either anterior chordal reattachment (n = 7) or posterior transposition (n = 7). Contractility was assessed using physiologic volume intercepts for end-systolic elastance, preload recruitable stroke work, and the relationship of the maximum rate of change of left ventricular pressure to the cnd-diastolic volume. The physiologic intercept for end-systolic elastance did not change after anterior or posterior MVR, but increased from 60 ± 14 mL before MVR to 72 ± 17 mL with conventional MVR (p < 0.002), indicating impaired left ventricular contractility. Similarly, the physiologic intercept for preload recruitable stroke work and the relationship of the maximum rate of change of left ventricular pressure to the end-diastolic volume increased 22% ± 13% and 28% ± 13%, respectively, after conventional MVR, but neither changed after anterior or posterior MVR. Although the end-diastolic pressure-volume relationship did not change with either chordal-sparing technique, its slope increased 98% ± 73% after conventional MVR (p < 0.008). Thus, although chordal preservation maintained better systolic and diastolic function, there was no substantial difference between the results of the anterior and posterior chordal-sparing techniques in this model.