Neurocognitive deficit following mitral valve surgery.

OBJECTIVE Neurocognitive deficit is an important complication in patients undergoing open heart surgery. The aim of this prospective, contemporary study was to objectively measure neurocognitive brain function following mechanical mitral valve replacement and mitral valve repair. METHODS Forty consecutive, unselected patients (mechanical valve replacement n=20, mean age 65+/-14; valve repair n=20, mean age 64+/-7, P=0.896) entered this prospective, contemporary study. Neurocognitive function was objectively measured by means of P300 auditory evoked potentials (peak latencies, ms) and two standard psychometric tests (Mini Mental State Examination, Trailmaking Test A (TTA)), preoperatively, 7 days and 4 months postoperatively. RESULTS Before operation, neurocognitive brain function was comparable in both patients groups (mechanical valve replacement versus valve repair: P300 potentials 374+/-25 versus 378+/-46 ms; P=0.791 and TTA 57+/-15 versus 54+/-10 s; P=0.552). Following mechanical valve replacement, neurocognitive function continuously worsened (7 day-follow-up: P300 potentials 392+/-28, P=0.001 versus preop and TTA 65+/-17, P=0.0001; 4-month follow-up: P300 potentials 406+/-39, P=0.0004; TTA 69+/-17, P=0.0001). Interestingly, neurocognitive brain function was unaffected in patients undergoing valve repair (7-day follow-up: P300 potentials 386+/-40, P=0.890 versus preop and TTA: 53+/-10, P=0.644; 4-month follow-up: P300 potentials 374+/-36, P=0.166 and TTA 54+/-11, P=0.147). At 4-month follow-up, patients with mechanical prostheses performed worse as compared to valve repair (P300 potentials: P=0.024; TTA P=0.014). CONCLUSION As shown by P300 auditory evoked potentials and Trailmaking Test A, there is marked neurocognitive damage related to mechanical valve replacement, whereas mitral valve repair does not affect neurocognitive function. This finding supports the beneficial effect of mitral valve repair over mechanical valve replacement in the decision-making tree of borderline cases, which are suitable for both types of procedure.