INTRODUCTION

Left atrial (LA) enlargement is commonly witnessed with cardiac valve disease. Increased LA volume is associated with atrial fibrillation (AF) and the risk of thrombus formation^[1]. Excessive LA enlargement can lead to bronchial compression or hoarseness of voice due to compression of the recurrent laryngeal nerve^[2]. Giant left atrium is a rare condition defined by a LA diameter > 65 mm and is often associated with long-standing rheumatic mitral stenosis^[3]. A giant left atrium may cause respiratory dysfunction and hemodynamic disturbance postoperatively. While operating for correction of mitral valve diseases, enlarged left atriums were reduced surgically^[4]. Some surgical groups recommend routine LA reduction with only a modest increase in its size and without any clinical signs of LA enlargement^[5]. There are no standard recommendations for this procedure based on the size of the LA^[4]. Most surgeons fix the mitral valve and do little to an oversized left atrium. Others occlude the LA appendage^[6]. We believe that a good proportion of surgeons think that successful mitral valve surgery alone will result in the eventual remodeling of the left atrium and size reduction. Risk of excessive bleeding, increased cardiopulmonary bypass (CPB) time, and unclear surgical efficacy raise many questions about the optimal approach to giant LA reduction^[7].

The purpose of this report is to assess the results of concomitant LA reduction with valve replacements in the treatment of heart valve diseases.

METHODS

RESULTS

Baseline and perioperative characteristics between the folded group and the unfolded group were showed in Table 1. Procalcitonin in the folded group was lower than in the unfolded group (0.06±0.12 vs. 0.15±0.2, P=0.003). LA max. diameters had no significant difference between the folded group and the unfolded group (88.2±29.9 vs. 81.7±16.9, P=0.13). In the folded group, there were 41 MVR, 18 double valve replacements (DVR), and three tricuspid valve replacements (TVR). In the unfolded group, there were 54 MVR and 17 DVR. The concomitant procedures of TVP in the folded group were similar to those in the unfolded group (54 vs. 53, P=0.084). There were more complications in the unfolded group including pneumonia, sepsis, MODS, LCOS, and CRRT. The ICU stay time and mechanical ventilation time had no significant differences. There were five deaths in the folded group and 25 deaths in the unfolded group (P<0.001). The ROC curve of LA max. diameter predicting death was analyzed in the folded group (Figure 1). The area under the curve was 87.8%, and the P-value was 0.005. The cutoff point of LA max. diameter was 96.5 mm.

Table 1 - Baseline and perioperative characteristics.

Fig. 2 - Receiver operating characteristic (ROC) curve. ROC curve for left atrial max. diameter predicting mortality. The cutoff point of left atrial max. diameter was 96.5 mm. AUC=area under the curve.

Download

Fig. 2 - Receiver operating characteristic (ROC) curve. ROC curve for left atrial max. diameter predicting mortality. The cutoff point of left atrial max. diameter was 96.5 mm. AUC=area under the curve.

According to the cutoff point of LA max. diameter, a subgroup analysis was performed in the folded group. The 63 patients were divided into the small (S) group (LA max. ≤ 96.5 mm, n=39) and the large (L) group (LA max. > 96.5 mm, n=24). Baseline characteristics were showed in Table 2. LA max. diameters in the S group were significant smaller than in the L group (69.6±18.5 vs. 118.3±17.5, P<0.001). The CPB times were shorter in the S group than in the L group (P=0.04). But the ACC time had no significant difference. The ICU stay time in the S group was significantly shorter than in the L group (2.2±1.0 vs. 6.0±8.2, P=0.032). Mechanical ventilation time in the S group was 17.2±11.0 hours, while it was 96.8±159.2 hours in the L group (P=0.023). AKI and LCOS in the S group were fewer than in the L group (P<0.05). All the five deaths occurred in the L group (P=0.013).

Table 2 - Subgroup analysis (S group: left atrial max. diameter < 96.5 mm; L group: left atrial max. diameter ≥ 96.5 mm).

We performed a stratified analysis for sex in both groups, and the results were showed in Table 3. It was found that the serum creatine was higher in male patients than in female patients in both groups. LA diameters had no significant differences between the two groups. In the unfolded group, more patients died in the female subgroup than in the male subgroup. Other characteristics had no significant differences in the stratified analysis.

Table 3 - Stratified analysis for sex in both groups.

Logistic regression for death has been performed in all patients (Table 4). Left atrium unfolded, left atrial max. diameter, CPB time, and mechanical ventilation time were the risk factors of death. Therein, the odds ratio of the unfolded left atrium was 22.72.

Table 4 - Results of logistic regression for death.

DISCUSSION

Our results demonstrated that LA reduction concomitant with valve replacement was safe and effective. The left atrium enlarged usually in rheumatic heart disease, especially in mitral valve disease, because of increased LA pressure and volume. The giant left atrium was known as the independent risk factor of AF. AF would decrease the cardiac output and cause atrial embolism^[8]. The giant left atrium could compress the trachea, bronchus, left ventricle, lung, esophagus, and recurrent laryngeal nerve surrounding. At the same time, the giant left atrium meant a longer time of illness and serious valves diseases. It predicted bad outcomes or more complications^[4]. Kawazoe^[9] reported that among patients who underwent valve operation but a giant left atrium was left free, 70% presented LCOS, 40% respiratory failure, and 40% died. These complications and mortality would decrease if valve operation was combined with LA reduction. Therefore, positive LA reduction was suggested in the course of valve replacements or plasty. In the present study, there were 25 deaths in the unfolded group and five deaths in the folded group. Pneumonia, sepsis, MODS, LCOS, and CRRT were significantly fewer in the folded group than in the unfolded group. These results were similar to other centers’ results^[5,7,10-15]. It might suggest that LA reduction could decrease mortality and incidence of complications.

There were several methods of plication^[4,14,16-18]. The left atrium might enlarge forward different directions. Different effects resulted from performing different atrial plication parts. For example, para-annular plication of LA wall among the mitral valve could mitigate the compression to the left ventricle and esophagus. Plication of interatrial septum could mitigate the compression to the right atrium. Therefore, according to the shape of the left atrium, we have chosen different kinds of plication to reduce LA volume. We suggest that the plication should be performed from inside of the left atrium. A suture from outside the left atrium might affect the pulmonary veins reflux. At the same time, it might easily cause atrial embolism for the rough intima. Some surgeons performed the left atrium plications when the heart was beating from outside the left atrium. We personally thought that was not suitable because it could result in more complications and bad clinical outcomes such as postoperative bleeding and mortality. Nevertheless, cardiac autotransplantation^[19,20] was thought to be the most radical technique of giant LA reduction which allowed maximally reducing all parts including interatrial septum. But the relative complexity and prolonged operative time made cardiac autotransplantation to be a technically demanding procedure and recommended only in case with extreme range of LA enlargement^[1]. Therefore, there wasn’t a sole standard plication method for giant left atrium. We could choose suitable plication methods according to the geometry of the giant left atrium.

AF was often concomitant in patients who had mitral valve disease and enlarged left atrium. An enlarged left atrium was a well-known risk factor for ablation failure of AF. Kasemsarn^[7] had performed mitral valve surgery concomitant with AF radiofrequency ablation and found that the preoperative LA diameter > 50 mm was the predictor of recurrence of AF. Kim^[10] had found that surgical AF ablation without LA volume reduction was inappropriate for patients with a giant left atrium and AF. Sunderland^[21] suggested that patients with an enlarged (≥ 55 mm) LA who were at risk of failing to obtain sinus conversion after a standard maze procedure might derive benefit from concomitant atrial reduction surgery using either a tissue excision or a tissue plication technique. In the present study, we did not perform AF radiofrequency ablation because of the giant left atriums and the patients’ willingness.

Giant left atrium was the end product of severe and prolonged pressure and volume overload, occurring mainly during mitral insufficiency, stenosis, and, rarely, in mitral valve prolapse alone^[4]. With the time going on, the left atrium could become larger and larger. Progressive LA dilatation might reflect the severity and duration of mitral valve disease^[22]. At the same time, some remodeling happened on the LA wall^[23]. Therefore, LA volume might be one of the risk factors for mitral valve disease. We have made the ROC curve for the LA max. diameter predicting death and found that the cutoff point of LA max. diameter was 96.5 mm (Figure 1). The patients of the folded group were divided into two subgroups according to the cutoff point. It was found that the L group had more complications and mortality than the S group. It meant that the enlargement of left atrium could predict the prognosis of patients, even though the LA reduction had been done. According to the results of logistic regression, there were four risk factors of death. The odds ratio of unfolded LA was 22.72. Therefore, the LA reduction procedure in such patients was strongly recommended.

CONCLUSION

Surgical LA reduction concomitantly with valve replacements could decrease mortality and were safe and effective in giant left atrium patients.