Modifications of the Cox-Maze III procedure.

Abstract

The extended operative time needed for surgery with complicated atrial incisions may preclude application of the Cox-Maze III procedure (CM-III) as a concomitant operation. And after the CM-III, left atrial (LA) contraction has been reported to recover in reduced magnitude compared with right atrial (RA) contraction. To decrease operative time, we have modified the CM-III (modification I) by: obliterating the LA appendage instead of excising it; cryoablating the bridge between the LA appendage and margin of the pulmonary vein encircling incision; extending the lateral incision of RA onto the RA appendage without excising it, and extending the incision more inferiorly toward the inferior vena cava; and omitting the T-incision of RA. We compared the clinical results of the conventional CM-III (group 1, n = 18) with those of the modified CM-III group (group 2, n = 23) performed in patients with rheumatic mitral valve (MV) disease. To enlarge the contractile area of the LA, we modified the CM-III to encircle the right and left pulmonary veins separately (modification II), and compared the LA contractilities of the conventional CM-III (group A, n = 15) with those of the second modification (group B, n = 9). Modification I: Mean aortic cross-clamp (ACC) times (135 +/- 29 versus 104 +/- 18 minutes, p < 0.005) and cardiopulmonary bypass (CPB) times (240 +/- 33 versus 185 +/- 42 minutes, p < 0.001) were significantly decreased in group 2 compared with those in group 1. In group 1, sinus rhythm was restored in 16 patients (88.9%). RA contractility was demonstrated in 100% of patients with sinus rhythm (16 of 16) and LA contractility in 75% (12 of 16) in the latest follow-up echocardiography. In group 2, sinus rhythm was restored in 21 patients (91.3%). RA contractility was demonstrated in 100% of patients with sinus rhythm (21 of 22) and LA contractility in 76.2% (16 of 21). Modification II: Mean ACC times were increased in group B compared with group A (133 +/- 32 versus 172 +/- 39 minutes, p = 0.02). The A velocities at LA contraction and the ratio of atrial contraction to peak early diastolic filling velocity (A/E ratio) of the trans-mitral flow were 0.14 +/- 0.20 m/sec and 0.23 +/- 0.11 in group A, and 0.58 +/- 0.33 m/sec and 0.47 +/- 0.19 in group B, respectively, both showing a significant increase in group B compared with group A (p < 0.05). Our first modification of the CM-III showed comparable sinus conversion rates and incidence of atrial contractility restoration with significantly shorter ACC and CPB times than the conventional CM-III. The second modification of the CM-III significantly increased the LA contractility when compared with the conventional CM-III, although the second modification required a longer ACC time.