Surgical treatment of cardiac lipoma: 20 years’ experience in a single center

Abstract

Background: 

Primary cardiac lipoma is very rare, and no consensus has been developed regarding its ideal treatment strategy. This study reviewed the surgical treatment of cardiac lipomas in 20 patients over 20 years.

Methods: 

Twenty patients with cardiac lipomas were treated at Fuwai Hospital, National Center for Cardiovascular Diseases, Chinese Academy of Medical Sciences and Peking Union Medical College from January 1, 2002, to January 1, 2022. The patients’ clinical data and pathological reports were retrospectively analyzed, and the follow-up with a range of 1 to 20 years was conducted.

Results: 

The cardiac lipomas were located in the right atrium (RA) or superior vena cava (SVC) in seven patients (35%) (atrium in six patiets and SVC in one patient), left ventricle in eight patients (40%) (left ventricular chamber in four patients and left ventricular subepicardium and myocardium in four patients), right ventricle in three patients (15%) (right ventricular chamber in one patient and right ventricular subepicardial layer and myocardium in two patients), subepicardial interventricular groove in one patient (5%), and pericardium in one patient (5%). Complete resection was achieved in 14 patients (70%), including seven patients with lipomas in the RA or SVC. Incomplete resection occurred in six patients (30%) with lipomas in the ventricles. No perioperative deaths occurred. Long-term follow-up was conducted for 19 patients (95%), including two (10%) who died. Both patients who died had lipomas incompletely resected due to ventricles involvement, and preoperative malignant arrhythmias persisted post-operatively.

Conclusions: 

The complete resection rate was high, and the long-term prognosis was satisfactory in patients with cardiac lipomas that did not involve the ventricle. The complete resection rate was low in patients with cardiac lipomas in ventricles; and complications, including malignant arrhythmia, were common. Failure of complete resection and post-operative ventricular arrhythmia are correlated with post-operative mortality.

Introduction

Primary cardiac tumors are extremely rare, with an incidence of 0.17% to 0.19% according to autopsy reports.^[1] Approximately 75% of primary cardiac tumors are benign, and 25% are malignant. The most common benign cardiac tumor is the cardiac myxoma, which accounts for 50% of benign tumors, followed by cardiac papillary fibroma and cardiac lipoma. Cardiac lipomas account for 2% to 8% of benign tumors, and the detection rate in the general population is <0.1% according to autopsy reports. Cardiac lipomas can originate in the endocardium, epicardium, myocardium, or pericardium and occur in any part of the heart, though the right atrium (RA) and left ventricle (LV) are the most common sites.^[2] Cardiac lipomas are often composed of mature adipocytes and have intact fibrous capsules.^[3] A preoperative diagnosis of cardiac lipomas can be made using a wide range of clinical imaging methods.^[4] While the cause of lipomas remains unclear, a previous study suggested that lipomas are associated with genetic abnormalities of chromosome 12.^[5]

Most patients with cardiac lipomas are asymptomatic, and tumors are often incidentally found during health check-ups. Depending on the location and size of the lipoma, symptoms of compression or obstruction, such as valvular dysfunction, congestive heart failure, and dyspnea, may occur.^[6,7] Arrhythmia due to compression has also been reported.^[8,9] When a cardiac lipoma becomes symptomatic, surgery to relieve symptoms and prevent disease progression is the preferred treatment method. However, no consensus regarding treatment strategies for asymptomatic patients has been reached.^[10,11] Most reports of cardiac lipomas are case studies. Few studies including large numbers of patients with cardiac lipomas have been reported, and the follow-up duration is short in most reports. The long-term post-operative outcomes of surgical treatment on cardiac lipomas remain unclear. Elbardissi et al^[6] summarized the clinical outcomes of patients with various types of cardiac tumors who underwent surgical treatment at their hospital during a 48-year-period. Only 12 patients had cardiac lipomas, and the long-term survival in these patients was worse than that of patients with cardiac myxomas. Due to the low clinical incidence of cardiac lipomas, the clinical characteristics have not been fully clarified. This study aimed to elucidate the clinical characteristics of cardiac lipomas and the long-term post-operative outcomes of patients who underwent surgery for cardiac lipomas via a detailed analysis of a relatively large group of patients treated at a single center.

Methods

Patients

From January 2002 to January 2022, 1164 consecutive patients underwent cardiac tumor resection at Fuwai Hospital, National Center for Cardiovascular Diseases, Chinese Academy of Medical Sciences and Peking Union Medical College, and 20 (2%) were diagnosed with lipomas by post-operative pathological examination. This study retrospectively analyzed the data of these 20 patients. Surgical reports and post-operative pathological results were used to confirm the patients’ eligibility for this study. Unlike previous study, patients with lipomatous hypertrophy of the interatrial septum were excluded in this study.^[12] Patients’ age, sex, diagnosis, complications, and laboratory results were extracted from the electronic medical records. The patients’ clinical records were reviewed for demographic information, clinical history, surgical reports, pathological reports, computerized multi-lead electrocardiograms, 24-h Holter monitoring records, and preoperative and post-operative imaging data including echocardiograms, magnetic resonance images [Figure 1A], cardiac computed tomography (CT) images, coronary angiograms, and cardiac positron emission tomography (PET) images. The accuracy of the patient information extraction was verified using double-entry techniques. This study was approved by the Ethics Committee of Fuwai Hospital, National Center for Cardiovascular Diseases, Chinese Academy of Medical Sciences and Peking Union Medical College (No. 2020-1360) and conducted according to the principles of the Declaration of Helsinki. The ethics committee waived the requirement of informed consent.

Figure 1: (A) Preoperative MRI of cardiac lipoma patient; arrow shows encapsulated mass. (B) Intraoperative images of cardiac lipoma patient. (C) Tumor size measurement. (D) Histological examination shows resected mass mostly comprising mature adipocytes (hematoxylin and eosin staining; original magnification × 200). MRI: Magnetic resonance imaging.

Surgical technique

The surgery was performed under general anesthesia and endotracheal intubation. Before the operation, transesophageal ultrasonography was used to confirm the location of the tumor and its relationships with surrounding tissues and structures. The surgery was performed at 32°C or normal temperature with and without cardiopulmonary bypass (CPB) in 15 and 5 patients, respectively. A median sternotomy was performed in all patients; then, the positions of the tumor and tumor pedicle and their relationships with important surrounding structures, such as coronary arteries and heart valves, had been carefully explored before the scope and method of the surgical resection were confirmed [Figure 1B]. A complete resection of the tumor tissue was performed as completely as possible. Cytoreduction was used when a complete resection could not be achieved. Tissue defects after tumor removal were repaired using autologous pericardium or polyester patches. Besides, abnormal structures and functions of the heart valves and other cardiac malformations were repaired during the surgery, and patients with coronary heart disease simultaneously underwent bypass grafting. After surgery, the size of the tumors was measured [Figure 1C], the pathological examination was performed to reveal the histological characteristics of the tumor [Figure 1D], and transesophageal echocardiography was also performed to evaluate the structure and function of the heart.

Follow-up

Telephone follow-up, face-to-face interviews, and supplementary follow-up data obtained from the patients’ outpatient records were used to extract data regarding adverse events after discharge. Adverse events included death and cardiovascular and cerebrovascular events, such as stroke, myocardial infarction, re-hospitalization for cardiac causes or pacemaker implantation, and secondary cardiac surgery. The patients’ post-operative symptoms and medication information were also obtained using structured follow-up forms. Following surgery, 19 patients (95%) were followed up to January 2022 or death. The median follow-up duration was 10 years (range, 1–20 years).

Statistical analysis

Continuous variables are presented as means ± standard deviations or medians with upper and lower quartiles. Categorical variables are presented as numbers and percentages. The Fisher's exact test was used to analyze the difference between the groups. Statistical significance was set at two-tailed P < 0.05.

Results

The basic characterics of enrolled patients were shown in Table 1. The mean age of patients at the time of surgery was 41.4 ± 14.6 years, and 55% (n = 11) of the patients were males. The primary symptoms were shortness of breath in five patients (25%), palpitation in six patients (30%), chest pain in four patients (20%), and dizziness, blackout, or syncope in three patients (15%). Eight patients (40%) were asymptomatic, and their cardiac lipomas were found incidentally by routine echocardiographic examination during health check-up. According to the New York Heart Association Functional Classification, seven (35%), 12 (60%), and one (5%) patient was classified as having class I, II, and III cardiac functions, respectively. Ventricular tachycardia (VT) occurred in four patients (20%), and two patients (10%) had diabetes. Hypertension, history of coronary heart disease, hyperlipidemia, thyroid cancer, teratoma, atrial septal defect, hypothyroidism, and mitral insufficiency were detected in one patient (5%), respectively. All 20 patients underwent preoperative echocardiography, and cardiac lipoma was diagnosed preoperatively in three patients (15%) by echocardiography. Eight patients (40%) underwent preoperative CT, and six (75%) were diagnosed with cardiac lipoma by CT. Thirteen patients (65%) underwent routine cardiac magnetic resonance imaging (MRI), and twelve (92%) were diagnosed with cardiac lipoma by MRI. PET was conducted preoperatively in two patients (10%), though neither was diagnosed with cardiac lipoma by PET. The largest diameter of the tumor was >5 cm in 14 patients (70%), and 12 patients of which (86%) had preoperative symptoms. Post-operative pathology confirmed the diagnosis of cardiac lipoma in all 20 patients (100%), and showed resected mass mostly comprising mature adipocytes [Figure 1D].

Table 1 - Characteristics of patients with cardiac lipoma (n = 20).

Items	Values
Gender (male)	11 (55)
Age (years)	41.4 ± 14.6
BMI (kg/m2)	23.7 ± 3.7
Comorbidities
Hypertension	1 (5)
Hyperlipidemia	1 (5)
Diabetes	2 (10)
History of coronary artery disease	1 (5)
Thyroid cancer	1 (5)
Teratoma	1 (5)
Atrial septal defect	1 (5)
VT	4 (20)
Hypothyroidism	1 (5)
Mitral insufficiency	1 (5)
Symptoms
Asymptomatic	8 (40)
Palpitations	6 (30)
Shortness of breath	5 (25)
Dizziness, blackout, or syncope	3 (15)
Chest pain	4 (20)
LVEF >55%	20 (100)
Hemoglobin (g/L)	136.5 ± 16.1
NYHA heart function classification
Class I	7 (35)
Class II	12 (60)
Class III	1 (5)
Examinations
Echocardiography	20 (100)
CT	8 (40)
MRI	13 (75)
PET	2 (10)

Data are shown as number (%) or mean ± standard deviation. BMI: Body mass index; CT: Computed tomography; LVEF: Left ventricular ejection fraction; MRI: Magnetic resonance imaging; NYHA: New York Heart Association; PET: Positron emission tomography; VT: Ventricular tachycardia.

According to preoperative imaging, the cardiac lipomas were located in the RA or superior vena cava (SVC) in seven patients (35%) (atrium in six patients and SVC in one patient), LV in eight patients (40%) (left ventricular chamber in four patients and left ventricular subepicardium and myocardium in four patients), right ventricle in three patients (15%) (right ventricular chamber in one patient and right ventricular subepicardium and myocardium in two patients), subepicardial interventricular groove in one patient (5%), and pericardium in one patient (5%). The tumor was located in a cardiac chamber in 11 patients (55%), in the subepicardium or myocardium in eight patients (40%), and in the pericardium in one patient (5%). In six patients (30%), the tumors were observed to have irregular shapes and unclear boundaries with invasion. Complete capsules and no invasion were observed in 14 patients (70%). The clinical characteristics of patients with different locations of cardiac lipoma were shown in Table 2.

Table 2 - Perioperative characteristics of patients with cardiac lipomas stratified by tumor locations.

	Tumor location
		Subepicardial and myocardium (n = 8)				Cardiac chamber (n = 11)
Items	Pericardium (n = 1)	SVC (n = 1)	LV (n = 4)	IVG (n = 1)	RV (n = 2)	RA (n = 6)	LV (n = 4)	RV (n = 1)
Preoperative symptoms	0 (0)	0 (0)	3 (75)	1 (100)	2 (100)	3 (50)	2 (50)	1 (100)
Preoperative FPVC	0 (0)	0 (0)	2 (50)	1 (100)	1 (50)	0 (0)	1 (25)	0 (0)
Preoperative VT	0 (0)	0 (0)	3 (75)	0 (0)	1 (50)	0 (0)	0 (0)	0 (0)
LDT >5 cm	1 (100)	1 (100)	3 (75)	1 (100)	2 (100)	3 (50)	2 (50)	1 (100)
Tumor invasion	0 (0)	0 (0)	3 (75)	1 (100)	0 (0)	0 (0)	1 (25)	1 (100)
Incomplete resection	0 (0)	0 (0)	3 (75)	1 (100)	0 (0)	0 (0)	1 (25)	1 (100)
Ventricular involvement	0 (0)	0 (0)	4 (100)	1 (100)	2 (100)	0 (0)	4 (100)	1 (100)
Death during FU	0 (0)	0 (0)	1 (25)	0 (0)	0 (0)	0 (0)	1 (25)	0 (0)
Post-operative symptoms	0 (0)	0 (0)	2 (50)	1 (100)	0 (0)	3 (50)	1 (25)	0 (0)
Post-operative VT	0 (0)	0 (0)	1 (25)	0 (0)	0 (0)	0 (0)	1 (25)	0 (0)
Loss to FU	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	1 (25)	0 (0)

Data are presented as number (%). FPVC: Frequent premature ventricular contractions; FU: Follow-up; IVG: Interventricular groove; LDT: Largest diameter of the tumor; LV: Left ventricle; RA: Right atrium; RV: Right ventricle; SVC: Superior vena cava; VT: Ventricular tachycardia.

Five patients (25%) who underwent surgery without CPB had pericardial or subepicardial lipomas. CPB was used in fifteen patients (75%), including seven (35%) in whom the right atrial approach was used, three (15%) in whom the right atrial and atrial septal double approaches were used, three (15%) in whom the left ventricular approach was used, one (5%) in whom the right atrial and right ventricular double approaches were used, and one (5%) in whom the right atrial, atrial septal, and left ventricular apical triple incision approaches were used.

Complete resection was achieved in 14 patients (70%), including seven patients with lipomas in the RA and SVC. Incomplete resection occurred in six patients (30%) with lipomas in the ventricles [Table 2]. Cardiac patch reconstruction was performed in five patients (25%) using autologous pericardium or polyester patches, including one patient (5%) who underwent repair of the SVC and the top of the left atrium and one patient (5%) who underwent mattress suturing of the left ventricular wound (four stitches using 1-0 prolene as well as mattress and continuous sandwich techniques). There were not perioperative complication such as deaths, major complications, or long-term tumor recurrence, in all patients. Surgical related data are presented in Table 3.

Table 3 - Operative characteristics of patients who underwent cardiac surgery for cardial lipoma.

Items	Values
CPB	15 (75)
CPB time (min)	85.1 ± 53.7
Aortic cross-clamping time (min)	62.9 ± 43.3
CPB approaches
RA	7 (35)
LV	3 (15)
RA + atrial septum	3 (15)
RA + RV	1 (5)
RA + atrial septum + LV	1 (5)
LDT (cm)	5.0 (3.0, 6.0)
Concomitant surgeries
Coronary artery bypass grafting	1 (5)
Mitral repair	1 (5)
Tricuspid valve repair	1 (5)
Atrial septal defect repair	1 (5)
Incomplete resection	6 (30)
Mechanical ventilation time (h)	11.4 ± 5.0
Operative death	0 (0)
Preoperative symptoms	7 (35)

Data are shown as number (%), mean ± standard deviation or median (Q₁, Q₃).CPB: Cardiopulmonary bypass; LDT: Largest diameter of the tumor; LV: Left ventricle; RA: Right atrium; RV: Right ventricle.

During the follow-up, atrial fibrillation occurred in one patient (5%) at 1 month post-operatively and was converted to sinus rhythm using medication. Coronary stent implantation was performed in one patient (5%) at 8 years after tumor resection due to coronary heart disease. Two patients (10%) died during the follow-up period due to complications of the tumor involving the LV [Table 2], including one patient who had paroxysmal VT preoperatively that did not resolve post-operatively. Though this patient was treated with radiofrequency ablation at 6 months post-operatively, VT remained, and sudden death occurred at 3 years post-operatively. The other patient was enrolled in this study at the timing of secondary surgery due to cardiac lipoma recurrence 20 years after the initial surgery. This patient who had frequent ventricular premature contractions preoperatively and VT post-operatively died suddenly at 1 year after the second surgery.

The patients were divided into incomplete and complete resection groups based on the surgical reports. No deaths occurred among the 14 patients who underwent complete resection. Two patients (33%) who underwent incomplete resection died; and there was no statistically significant difference between the two groups (P = 0.143). Four (67%) and three patients (21%) experienced symptoms post-operatively in the incomplete and complete resection groups, respectively; and there was a statistically significant difference between the two groups (P = 0.012). Two patients (33%) in the incomplete resection group had ventricular arrhythmia (VA) after surgery, compared to that of no patient in the complete resection group, but the statistical difference between the two groups was not observed (P = 0.143).

The patients were divided into ventricular involvement and non-ventricular involvement groups, as detected via preoperative imaging and intraoperative exploration. Two (17%) of the twelve patients with ventricular involvement and none of the eight patients in the non-ventricular involvement group died (P = 0.648). Four patients (33%) with and three patients (38%) without ventricular involvement had post-operative symptoms (P = 0.774). Two patients (17%) in the ventricular involvement group had post-operative VT, compared to that of no patient in the non-ventricular involvement group (P = 0.648).

Discussion

Cardiac lipoma is a rare primary cardiac tumor. Maurer et al^[13] reported the first surgical removal of cardiac lipoma in 1952; however, few reports including a large number of patients have analyzed the characteristics, treatments, and prognoses of patients with cardiac lipomas.^[7] This study included 20 patients with cardiac lipomas who underwent cardiac surgery at our hospital. This is the first study on cardiac lipomas to include such a large number of patients. The presence of symptoms was related to tumor size and whether the subendocardial ventricular muscle was involved. Lipomas growing outside the ventricles were able to be completely removed, resulting in a favorable long-term prognosis. Complete removal of tumors located in the LV with invasive pathological manifestations was not possible, and these patients often experienced VA preoperatively. Failure of complete resection of the cardiac tumor may be related to the occurrence of adverse events post-operatively. Therefore, the preoperative symptoms and examination results can be used to preoperatively determine whether the tumor can be removed completely and predict post-operative outcomes.

Cardiac lipoma is a benign cardiac tumor composed of mature adipocytes. It often has a complete capsule, though tumors with no capsule have also been reported. Approximately 25% of lipomas are located in the endocardium, 25% in the myocardium, and 50% in a ventricular chamber.^[2] A previous literature review suggested that the atrial septum was the most common site of cardiac lipoma (39%), followed by the RA (18%), pericardium (13%), LV (12%), right ventricle (5%), ventricular septum (3%), and atrioventricular septum (1%).^[14] In this study, the tumors were located in the RA chamber in 30% of the patients, in the LV chamber in 20% of the patients, and in the RV chamber in 5% of the patients, while the remaining tumors were located in the endocardium and myocardium. No tumors were located in the left atrium in this study. The discrepancies on tumor location between the present study and previous study^[14] may be because lipomatous hypertrophy of the interatrial septum, which was excluded in this study, was classified^[12] as cardiac lipoma in the previous study.

Echocardiography is the most basic and important detection method for the diagnosis of cardiac tumors, as it is easy to use, cost-effective, and portable, and does not expose the patient to electromagnetic radiation.^[15,16] Cardiac lipomas typically appear as non-specific echoes on ultrasound with few blood vessels inside the tumor. Cardiac lipomas also appear as homogeneous echoes, without calcification and with significant hyperechoic masses and adipose tissue. The shape of the tumor may be irregular and without significant tumor pedicle. Cardiac lipomas typically have a clear boundary, though the boundary may not be clear in tumors invading into the myocardium or other tissues.^[14,17] The position, shape, size, and real-time movement of the lesion can be easily monitored using ultrasound. However, the accurate histological features of the tumor cannot be determined using ultrasonography. Cardiac CT allows for a more accurate evaluation of the histological nature of the tumor and the degree of its invasion of adjacent organs and myocardium. Cardiac lipomas exhibit low attenuation and are characterized by a uniform appearance of the adipose tissue, similar to that of the subcutaneous or mediastinal fat, with a Hounsfield value of <50 on CT. As lipomas have a low density on CT, they are easily distinguished from the myocardium, thrombi, and other tumors, including myxomas.^[18] Cardiac CT has a good temporal and spatial resolution. When combined with the gating technology of electrocardiograms, more accurate three-dimensional reconstruction of the tumor can be achieved.^[19] Cardiac nuclear MRI (NMRI)allows for the more accurate diagnosis and differentiation of cardiac lipomas. Cardiac NMRI reveals the tissue composition of the tumor, the degree of invasion the cardiac structures, and the effects on hemodynamics.^[20] Due to the high resolution of T₁- and T₂-weighted scans, cardiac MRI allows for a precise evaluation of the nature of the tumor and the degree of invasion,^[17,21] and it can be used preoperatively to determine whether complete resection of the tumor can be achieved. In this study, 13 patients underwent cardiac MRI preoperatively, and two of them were determined to have invasive growth into the myocardium. Although involvement of the papillary muscles was detected intraoperatively in one patient who was not detected by MRI, cardiac MRI is important for the preoperative planning of the surgical treatment of cardiac lipomas.

Small-diameter cardiac lipomas may be asymptomatic and are often found incidentally. However, the tumor may become symptomatic as it grows.^[22] Tumors with subendocardial and myocardial involvement may cause compression symptoms. Intracardiac lesions may lead to dyspnea secondary to blood flow obstruction or syncope, arrhythmia, palpitation, and angina pectoris. Surgical resection was conducted in eight patients without symptoms in this study based on the assumption that cardiac lipomas would grow and cause symptoms or complications such as coronary artery compression, pericardial tamponade, malignant arrhythmia, outflow tract obstruction, or left ventricular aneurysm.^[23] Although the long-term prognosis of most asymptomatic lipomas is good, untreated cardiac lipomas may lead to sudden death.^[24]

A previous study regarding cardiac tumors in a large patient population^[10] reported six patients with cardiac lipoma, including four in the RA, one in the ascending aorta, and one in the LV. The tumors were removed completely without post-operative complications or death within 3 months in all six patients in the previous study. No post-operative VT was reported in the previous study, though the follow-up period was short. In this study, the median follow-up duration was 10 years. Two patients who had cardiac lipomas involving ventricles with myocardial invasion, preoperative VT, and incomplete tumor removal died during the long-term follow-up in this study. D'Souza et al^[25] suggested that the invasion of the tumor into the myocardium affects the electrical conduction system of the heart, leading to arrhythmia. In this study, four patients with tumors involving ventricles had preoperative VT, including two with post-operative VT who died. Six patients in this study had a myocardial invasion of the ventricular muscle, and complete resection could not be achieved in these patients. Two of them died. When the tumor involves ventricle and invades the myocardium, it may lead to post-operative VT and a poor prognosis. Patients in whom complete resection cannot be achieved may benefit from the implantation of a cardioverter defibrillator.

This study is limited by the small sample size that prohibits rigorous statistical analyses. However, it is the largest report regarding cardiac lipomas with the longest follow-up duration, to date.

In summary, complete resection can be achieved for cardiac lipomas without ventricular involvement when no myocardial invasion has occurred, and the long-term post-operative outcomes are satisfactory. In contrast, complete resection is difficult, and the long-term outcome is poor for patients with cardiac lipomas involving ventricles with myocardial invasion.

Acknowledgements

We gratefully acknowledge the dedicated efforts of the participants, and the clinical staff involved in data collection. We would like to thank Editage (www.editage.cn) for English language editing.

Conflicts of interest

None.