We analyzed the risk of postoperative stroke after mitral valve repair according to cardiac rhythm and coumadin therapy, to determine whether patients (pts) in sinus rhythm (SR) could be safely managed without coumadin.
We studied 561 pts who were discharged without mortality or perioperative neurological events. Mean age was 62±15 years, 51.5% male. Atrial fibrillation (AF) was present preoperatively in 31.4%. Preoperative NYHA class was III-IV in 52.0% pts; coronary artery disease was present in 37.4%. Previous cardiac surgery had been done in 5.5% pts. Mean ejection fraction was 54.5±13.1%. The pts were followed up to 15 yrs, mean 1.21±2.25 yrs.
Strokes occurred in 17 pts at a mean post-operative interval of 3.8 years, range 0.6 - 14.9 years. There were no strokes after isolated mitral valve repair in the first six mths after surgery. Of the 182 (32.8%) pts with AF, 101 were discharged on no coumadin and 5 had strokes (5.0%). Of the 81 pts with AF discharged on coumadin, 4 had strokes (4.9%). P=N.S. Of the 294 pts in SR discharged on no coumadin 5 (1.7%) had strokes while 2/53 (3.9%) on coumadin had strokes. P=0.05. Pts with AF and no coumadin had a relative risk (RR) of stroke of 2.9 vs. pts in SR and on no coumadin. Cox regression multivariate analysis showed that SR preoperatively was the strongest predictor of freedom from stroke (p=0.04).
These data support the paramount importance of sinus rhythm as a predictor of freedom from stroke: 7/345 (2.0%) pts in SR had strokes vs. 9/182 (5.0%) in AF. Throughout the duration of this study, pts in sinus rhythm discharged on no coumadin rarely had strokes (5/294, 1.7% total strokes).
We believe these data support our policy of avoiding post-operative coumadin completely for pts in SR and emphasize the need for coumadin for pts in AF. These results also emphasize the benefit of performing mitral valve repair before the onset of AF.
G.M. Lawrie, None.