Comparison of 2 approaches for left atrial myxoma excision: Outcomes and safety

Abstract

Objective

Cardiac myxomas are the most common primary benign cardiac tumors, with approximately 75% located in the left atrium (LA). The optimal surgical approach for LA myxoma excision remains controversial. This study aims to compare the uniatrial and biatrial approaches in terms of safety and outcomes.

Methods

We retrospectively reviewed 157 consecutive patients diagnosed with LA myxoma over a 24-year period. Patients were divided into uniatrial (LA group) and biatrial groups on the basis of the surgical approach to compare outcomes.

Results

There was no statistically significant difference between the 2 groups in mean age (P = .61), gender distribution (P = .72), clinical presentation (P = .071), New York Heart Association classification (P = .97), preoperative and postoperative rhythm (P = .83 and P = .20, respectively), concomitant procedures (P = .63), overall postoperative complications (P = .32), re-exploration for bleeding (P = 1), 30-day mortality (P = 1), in-hospital stay (P = .08), long-term survival (P = .42), and recurrence rate (P = .38). However, the mean cardiopulmonary bypass time and aortic crossclamp time were significantly longer in the biatrial group (cardiopulmonary bypass: 81 vs 72 minutes, P = .05; crossclamp: 55 vs 48 minutes, P = .05). Kaplan-Meier survival analysis showed no significant difference in short-, mid- (5-year), and long-term (>10-year) survival rates. Eight cases initially planned for LA approach were converted to biatrial because of technical difficulty.

Conclusions

To the best of our knowledge, this is the largest single-center study on this topic. Both uniatrial and biatrial approaches provide safe and effective resection of left atrial myxomas with comparable long-term outcomes. The biatrial approach offers improved exposure but at the expense of longer operative times. These findings support an individualized surgical strategy on the basis of tumor characteristics and intraoperative exposure.

Uniatrial and biatrial approaches achieve similar outcomes in left atrial myxoma surgery.

Central Message.

Left atrial myxoma can be safely excised via either approach; the direct approach reduces operative times without compromising outcomes.

Perspective.

There is no standard approach for left atrial myxoma excision, and selection depends on surgeon preference and tumor complexity. In this largest single-center study, direct and biatrial approaches achieved equivalent outcomes; the direct route offered shorter operative times, supporting individualized approach selection.

Cardiac myxomas are the most common primary benign cardiac tumors, with 75% located in the left atrium.1, 2, 3 The estimated incidence is 0.03%, occurring more frequently in female than male patients.^4,5 Although benign, myxomas can cause serious complications, including neurologic events, which are noted in up to 25% of patients.² Cardiac myxomas can be either sporadic (90%) or familial, the latter being associated with the Carney complex due to PRKAR1A gene mutations.^2,6 The familial type accounts for approximately 5% of cases and is characterized by male predominance, younger age at presentation, and a greater recurrence rate.^2,5

Clinically, left atrial (LA) myxomas manifest with 3 groups of symptoms: hemodynamic consequences, systemic embolism, and constitutional symptoms. Interestingly, up to 46.3% of cases can be asymptomatic.^2,3 Echocardiography, which includes transthoracic echocardiography (TTE) and transesophageal echocardiography (TEE) approaches, is the first-line diagnostic method. Cardiac magnetic resonance imaging and/or computed tomography can be used as second-line modalities when diagnostic uncertainty exists.¹

Surgical resection is the treatment of choice for LA myxomas, with low mortality and recurrence rates of up to 1% and 0.5 cases per 1000 person-years, respectively.^3,5 In patients with Carney complex, recurrence rates can reach 22%, compared with 1 to 3% in sporadic cases.³ Recurrence is generally attributed to incomplete excision, tumor spillage, or new tumor focus activation.⁶ The purpose of this study is to compare the uniatrial and biatrial approaches for LA myxoma excision at our center over a 24-year period.

Methods

The data of the patients were retrospectively reviewed from February 2001 to August 2024; a total of 197 patients were operated for atrial tumors in our department. Of those, 157 patients had undergone surgical resection for LA myxoma. Right atrial tumors or any histopathology other than myxoma were excluded from the analysis. The study protocol was approved by the Ethics Committee of the University Hospital Southampton (number and date: SEV/0767, September 2, 2024). Informed patient consent was not required, because it had been obtained at the time of surgery.

Preoperative Assessment

All included patients were clinically assessed and investigated accordingly. Preoperatively, the diagnosis was established by TTE, TEE, or both. Intraoperatively, TEE was routinely used to inspect all 4 cardiac chambers. Coronary angiography or computed tomography coronary angiogram was performed if the patient was older than 40 years of age. Coronary assessment was performed selectively in patients younger than 40 years when risk factors or family history were present. In addition, routine laboratory investigations, including full blood count and blood biochemistry and C-reactive protein, were performed.

Surgical Technique

Once the diagnosis was established, all patients underwent operation as soon as possible. Median sternotomy approach was used for all patients. Cardiopulmonary bypass (CPB) perfusion was established through aortic and bicaval cannulations or femoral cannulations in redo cases. Minimal manipulation was performed before aortic crossclamp, taking in consideration the friability of this tumor to avoid emboli. Myocardial protection was achieved with antegrade cold blood cardioplegic arrest. In a combined procedure, LA myxoma has to be removed first. The patient were divided into 2 groups for surgical approaches comparison, which consisted of the uniatrial approach (LA approach) and biatrial approach, depending on surgeon preference and the tumor complexity.

The LA Approach

An interatrial groove was developed, and an LA incision was performed starting at the confluence of the right superior pulmonary vein and was extended in both directions parallel to the interatrial groove.

The Biatrial Approach

The biatrial approach in this study included transseptal (TS) and mainly superior transseptal (STS) approaches. The right atrium was incised parallel to the atrioventricular groove. In the TS approach, a vertical incision was created in the fossa ovalis and was extended in both directions whereas in the STS approach the fossa ovalis incision was extended to join the right atriotomy and then extended onto the roof of the LA behind the aorta.

Complete excision of the tumor with full-thickness resection of the base was performed for all patients to avoid any recurrence. In the biatrial approach, all cardiac chambers were inspected meticulously for any other tumor/residual fragments. An autologous or bovine pericardial patch was used to close the interatrial septum when the excision was wide, to avoid any tension to the septum; otherwise, it was closed by primary closure. Deairing and the remainder of the procedure were completed in the standard fashion. All resected tumors included in this study were diagnosed as benign LA myxoma.

Postsurgical Follow-up

Postoperatively, patients were followed up at throughout their stay in hospital by physical examination, continuous electrocardiogram monitoring if required and daily 12-lead electrocardiogram. All patient were checked by TTE before discharge and yearly after that. All patient continued their follow up at our outpatient department.

Statistical Analysis

Statistical analysis was performed using the SPSS, version 23 (IBM Corp) for Windows. The mean ± standard deviation of the data were calculated The suitability of continuous variables to normal distribution was tested by the Kolmogorov-Smirnov test and data were analyzed using the Student t test (parametric data) or Mann-Whitney U test (nonparametric data). The χ² test for categorical data. Kaplan-Meier method was performed for the long-term cumulative survival analysis.

Results

Table 1 presents the clinical and demographic data of the 2 patient groups. The study cohort consisted of 157 patients, of whom 115 (73.2%) underwent surgical resection via a biatrial approach (100 with an STS approach and 15 with a TS approach), whereas 42 patients (26.8%) underwent a uniatrial (LA) approach. Eight patients initially planned for the LA approach were converted intraoperatively to the biatrial approach and were analyzed accordingly within the biatrial group.

Table 1.

Demographic and clinical data and comparison between the LA group and the biatrial group data

Variables	All (N = 157)	LA approach (n = 42)	Biatrial approach (n = 115)	P value
Age, y	65 ± 12	64 ± 12	65 ± 12	.61
Gender, female	107 (68.2%)	27 (64.3%)	80 (69.6%)	.72
NYHA class				.97
I	95 (60.5%)	24 (57.1%)	71 (61.7%)
II	20 (12.7%)	6 (14.3%)	14 (12.2%)
III	36 (22.9%)	10 (23.8%)	26 (22.6%)
IV	7 (4.5%)	2 (4.8%)	5 (4.3%)
Preoperative rhythm				.83
Sinus	139 (88.5%)	37 (88.1%)	102 (88.7%)
Atrial fibrillation	18 (11.4%)	5 (11.9%)	13 (11.3%)
Permanent pacemaker	11 (7.0%)	2 (4.8%)	9 (7.8%)	.52
Crossclamp time, min	53 ± 28	46 ± 28	57 ± 28	.05
Cardiopulmonary bypass time, min	79 ± 37	68 ± 36	83 ± 36	.05
Combined procedures	42 (26.8%)	10 (23.8%)	32 (27.8%)	.63
CABG	18 (11.4%)
Valve	14 (8.9%)
Aorta	3 (1.9%)
Others	10 (6.4%)
Estimated blood loss, mL	320 ± 213	270 ± 172	337 ± 225	.20
Hospital length of stay, d	10 ± 6	9 ± 4	10 ± 6	.08
ITU length of stay, h	46 ± 61	36 ± 22	49 ± 68	.28
Postoperative mortality	1 (0.6%)	0	1 (0.9%)	1.00
Postoperative complications
Arrhythmia	68 (43.3%)	10 (23.8%)	40 (34.8%)	.20
Acute kidney injury	7 (4.5%)	2 (4.8%)	5 (4.3%)	.90
Infection	19 (12.1%)	1 (2.4%)	6 (5.2%)	.45
Permanent pacemaker insertion	11 (7.0%)
Low cardiac output	2 (1.3%)	1 (2.4%)	1 (0.9%)	.45
Bleeding requiring re-exploration	1 (0.6%)	0	1 (0.9%)	1.00
Neurologic event	2 (1.3%)	0	2 (1.7%)	1.00
Miscellaneous	1 (0.6%)	0	1 (0.9%)	1.00
Long-term mortality	43 (27.4%)	9 (21.4%)	34 (29.6%)	.42
Follow-up interval, mo	95 ± 77	106 ± 74	92 ± 78	.30
Recurrence	1 (0.6%)	0	1 (0.9%)	.38

ITU, Intensive therapy unit;LA, left atrium; NYHA, New York Heart Association; CABG, coronary artery bypass grafting.

There were 107 female patients (68.2%) and 50 male patients (31.8%), with a mean age of 65.3 ± 11.7 years at the time of surgery. The most common presenting symptom was dyspnea, reported in 58 patients (36.9%), followed by asymptomatic presentation in 41 patients (26.1%), chest pain in 14 patients (9.0%), constitutional symptoms in 8 patients (5.1%), and arrhythmias in 2 patients (1.3%). In addition, 33 patients (21.0%) presented with symptoms of systemic embolization. Most patients were in New York Heart Association (NYHA) functional class I at admission: 94 patients (59.9%) were class I, 20 (12.7%) were class II, 36 (22.9%) were class III, and 7 (4.5%) were class IV.

Combined procedures were performed in 42 patients (26.8%), as detailed in Table 1. Three patients (1.9%) required re-exploration for postoperative bleeding, and 11 patients (7.0%) required permanent pacemaker implantation. Postoperative complications occurred in 82 patients (52.2%), mainly arrhythmias (68 patients, 43.3%). These complications are outlined in detail in Table 1.

One patient (0.6%) experienced recurrence and had a known history of Carney complex. No recurrence was observed among patients with sporadic myxomas. Perioperative mortality was 0.6%, with 1 patient dying on postoperative day 5 due to a massive hemorrhagic stroke.

The mean follow-up period was 98.4 ± 77.3 months. Wide resection of the interatrial septum to achieve full-thickness margins was necessary in 51 patients (30.2%), for whom either autologous or bovine pericardial patches were used. The mean CPB time was 78 ± 36 minutes, and the mean aortic crossclamp time was 53 ± 28 minutes. The average in-hospital stay was 10 ± 6 days. All patients completed follow-up.

The overall survival rate was 68%, with 42 patients dying from various nonsurgical causes. Among these, 26 (62%) were octogenarians. Importantly, all patients survived for at least 1 year postoperatively. The Kaplan-Meier survival curve is shown in Figure 1.

Figure 1.

Overall survival.

To compare the outcomes and safety of the 2 surgical approaches, data from the LA and biatrial groups are shown in Table 1. There were no statistically significant differences between the 2 groups in terms of mean age (P = .61), gender distribution (P = .72), clinical presentation (P = .071), NYHA classification (P = .97), preoperative and postoperative rhythm (P = .83 and P = .20, respectively), concomitant procedures (P = .63), overall postoperative complications (P = .32), re-exploration for bleeding (P = 1), permanent pacemaker insertion (P = .52), 30-day morality (P = 1), in-hospital stay (P = .08), long-term survival (P = .42), and recurrence rate (P = .38).

However, the biatrial approach was associated with significantly longer CPB time (P = .05) and aortic crossclamp time (P = .05). The Kaplan-Meier survival analysis (Figure 2) showed no statistically significant differences between the 2 approaches in terms of short-, mid- (5 years), and long-term (>10 years) survival.

Figure 2.

Kaplan-Meier analysis demonstrated excellent early survival in both groups. At 1 year, survival was 97.6% (95% CI, 93.0-100) in the left atrial (LA) group and 98.1% (95% CI, 95.5-100) in the superior transseptal (STS) group. At 3 years, survival was 95.0% (95% CI, 88.5-100) for LA and 92.7% (95% CI, 87.6-98.1) for STS. Five-year survival was 92.1% (95% CI, 83.9-100) in the LA cohort compared with 82.1% (95% CI, 74.1-90.9) in the STS cohort. Ten-year survival was 80.7% (95% CI, 67.5-96.5) in the LA cohort compared to 66.9% (95% CI, 56.4-79.2).

In a multivariable Cox proportional hazards model adjusted for age, gender, preoperative rhythm, NYHA class and combined procedures, surgical approach was not associated with long-term mortality (hazard ratio, 1.58; 95% CI, 0.73-3.43, P = .25). Age was an independent predictor of late mortality (hazard ratio, 1.07; 95% CI, 1.04-1.10, P < .01). Gender, NYHA class, preoperative rhythm, and combined procedures were not significantly associated with mortality.

Firth logistic regression was constructed to identify independent predictors of postoperative arrhythmia, adjusting for age, gender, preoperative rhythm, NYHA class, combined procedures, cardiopulmonary bypass duration, and aortic crossclamp time. None of the variables were independently associated with postoperative arrhythmia. Surgical approach demonstrated a nonsignificant trend toward reduced arrhythmia in the biatrial group (odds ratio, 0.43; 95% CI, 0.17-1.00, P = .05).

Discussion

Myxomas are neoplasms of uncertain origin; however, most studies suggest they arise from subendocardial multipotent mesenchymal cells.⁴ Surgical excision is the standard of care for LA myxomas. The first successful excision was performed by Crafoord in 1954.^2,4 Although the urgency of surgery remains debated, early removal is generally recommended due to the tumor's potential for serious complications and the low risk of postoperative morbidity and mortality.^2,7

The optimal surgical approach for LA myxoma remains controversial. There is a growing preference in the literature for the biatrial approach.^7,8 In general, 2 main surgical strategies exist: the uniatrial (LA) approach and the biatrial approach. Guiraudon and colleagues introduced the STS approach, which allows easier access and manipulation of the fragile tumor, especially in small LA, and enables inspection of all 4 cardiac chambers.⁹ Kabbani and Cooley later popularized the biatrial approach in 1973.^4,10

The ideal approach should provide excellent exposure to minimize tumor manipulation, reduce the risk of fragmentation and recurrence, and allow for safe excision and reconstruction if required.⁸ Proponents of the LA approach cite advantages such as requiring only a single atrial incision, reduced bleeding risk, fewer arrhythmic complications, and a low recurrence rate. However, its limitations include restricted exposure, which may necessitate conversion to the biatrial approach. In addition, tumors in unfavorable locations may require excessive manipulation, potentially compromising safety.^7,8,11 That said, for small tumors located in the posterior LA, the uniatrial approach may be sufficient.^4,6

In contrast, the biatrial approach—especially the STS variant—offers excellent exposure, facilitating safe and thorough excision and enabling complete inspection of all 4 cardiac chambers. This makes it particularly useful in small, deep-seated LA and in redo cases.⁷ Moreover, complete and wide excision with clear margins, which is essential to minimize recurrence risk, can often be achieved more easily via the biatrial approach.⁴

Critics of the biatrial technique point to increased CPB and aortic crossclamp times, greater bleeding risk, and a greater incidence of early and late conduction disturbances.^7,8,11 The pathophysiology of these conduction disturbances is not fully understood but may involve injury to the sinus node artery or result from extensive atrial incisions and retraction.⁷ Interestingly, electrophysiological studies and heart transplant research suggest that preserving the sinus node artery is not essential to prevent postoperative arrhythmias.⁵

In our series, although the bleeding rate was greater in the biatrial group (1.8% vs 0%), this difference was not statistically significant. Similarly, the rate of permanent pacemaker implantation was greater in the biatrial group, but not significantly so, and no difference in the incidence of arrhythmic complications was observed between the 2 groups. However, CPB and crossclamp times were significantly longer in the biatrial group, which aligns with expectations due to the added complexity of closing 2 atrial incisions.

In this study, both the uniatrial and biatrial approaches were effective in achieving complete tumor excision, with comparable rates of postoperative morbidity, mortality, and long-term survival. However, in 8 cases, the LA approach proved inadequate, and the biatrial approach was adopted intraoperatively to facilitate tumor removal.

The LA (uniatrial) approach can be considered in selected patients with small tumors and adequate LA size. However, the biatrial approach has demonstrated superior efficacy and safety, offering better exposure and access, and remains closer to the ideal technique for complete and safe excision of LA myxomas. Although surgical approach selection is often considered dependent on the surgeon, our study provides one of the largest contemporary single-center comparisons of uniatrial and biatrial techniques. We highlight that both approaches achieve comparable survival and recurrence outcomes while the biatrial technique is associated with longer cardiopulmonary bypass and crossclamp times. These findings support an individualized surgical strategy based on tumor characteristics rather than a uniform approach.

Conflict of Interest Statement

The authors reported no conflicts of interest.

The Journal policy requires editors and reviewers to disclose conflicts of interest and to decline handling or reviewing manuscripts for which they may have a conflict of interest. The editors and reviewers of this article have no conflicts of interest.