SPONTANEOUS CARDIOVERSION AND MITRAL-VALVE REPAIR - A ROLE FOR SURGICAL CARDIOVERSION (COX-MAZE)

Objective: The objectives of this study are to describe: (1) The incid ence of change in pre-operative rhythm (cardioversion) with mitral val ve repair early and at 1 year's review after surgery (late). (2) The c haracteristics of those patients who remain in atrial fibrillation (AF ) or sinus rhythm (SR) at late follow up. (3) The characteristics of t hose patients whose rhythm is seen to change (cardiovert) from SR to A F, or AF to SR and to remain so at 1 year. In this way it is hoped to more clearly define those patients who would benefit from the combinat ion of mitral valve repair and surgical cardioversion (Cox-maze proced ure), Methods: Retrospective study was made of the case notes of all p atients undergoing mitral repair at our hospital during the 3 years be tween January 1st, 1991 and December 31st, 1993. Early (hospital disch arge) and late (1 year) post operative e.c.g, rhythm was compared to p re-operative e.c.g. rhythm. The study explored the association of card ioversion with pre-operative rhythm, patient age, aetiology of mitral valve lesion (mitral regurgitation or stenosis) and echo cardiographic estimations of left atrial size and left ventricular dimensions. Resu lts: Patients (89) underwent repair with a 30 day mortality of 2.2% (2 of 89). Of these, 55 were male with an average age of 65 +/- 12 years . Regurgitation was the valvular lesion in 93% and 18% were associated with coronary artery disease, 48 (55%) were in SR before surgery. Bot h deaths occurred in patients with AF as a pre-operative rhythm. Of th e 39 survivors originally in AF, only one was of recent onset (< 6 mon ths). The frequency of an enlarged left atrium (greater than or equal to 5.0 cm) was significantly greater in those with AF compared to SR ( P < 0.001). Atrial fibrillation was also associated with increasing ag e (P = 0.006) and increasing left ventricular end systolic diameter (L VESD; P = 0.018). Spontaneous cardioversion of pre-operative rhythm wa s common at the time of hospital discharge (AF to SR: 46% and SR to AF : 25%). At the 1 year review after mitral repair only 8 (21%) of those originally in AF were then in sinus rhythm. Eight (17%) of those orig inally in SR were in AF. A lower left ventricular end systolic diamete r (LVESD) was associated with spontaneous cardioversion of AF to SR by one year (P = 0.005). Similarly, patients originally in SR with a low er LVESD continued in SR. Those with a higher value were seen to cardi overt to AF (P < 0.05). Conclusions: Immediately prior to surgery the presence of AF was associated with a tendency to larger left atrial si ze, older age and a greater LVESD. Cardioversion was common for both p atients in AF (46%) and SR (25%) early following conservative mitral s urgery. The prevalence of late cardioversion was of a similar order in both those originally in AF (21%) and SR. (17%). The maintenance of, or cardioversion to SR seemed to be characterised only by the LVESD. T his analysis captures many of the problems of retrospective review. A multi-centre, prospective study is proposed to achieve the aim of an a ccurate formula predicting long standing cardioversion with mitral val ve surgery. Copyright (C) 1997 Elsevier Science B.V.