Primary cardiac tumors are rare neoplasms with an annual global incidence of less than 0.2% (1). Approximately 90% of these tumors are benign, the majority of which are cardiac myxomas (CMs) and papillary fibroelastomas (CPFEs). CMs arise from multipotent mesenchymal cells and consist of a myxoid matrix with an acid-mucopolysaccharide-rich stroma (2). Meanwhile, CPFEs are benign lesions derived from valvular endocardium with unknown pathogenesis. There is debate about whether CPFEs represent true neoplasms or reactive tumors; CPFEs are suggested to be hamartomas, neoplasms, or organizing thrombi, in which they originate from aggregates of microthrombi at sites of endothelial damage that evolve into a mass (3). Although generally identified postmortem, increasing utilization and resolution of echocardiographic modalities have shown that CPFEs may have a prevalence equivalent to CMs (4). Clinical presentation for these benign cardiac neoplasms varies and significantly depends on tumor location. Left-sided CMs present with obstructive and constitutional symptoms and systemic embolisms. CPFEs, especially those found on the mitral and aortic valves, are associated with thromboembolic disease (5, 6).

The long-term prognosis of CMs and CPFEs is good, with low recurrence rates often associated with either a relevant family history or an absence of negative margins on surgical excision (1, 7). There is a relative scarcity of understanding of the pathogenesis, clinical profile differences, tumor size markers, and long-term prognosis of CMs and CPFEs. Therefore, this institutional retrospective cohort study aims to provide a comprehensive characterization of the clinical course of patients diagnosed with CMs and CPFEs. In the study group, we emphasize potential prognostic markers, including neutrophil–lymphocyte ratio (NLR), and underline neurovascular manifestations.

In this study, we present clinical, echocardiographic, and laboratory data of 55 patients with benign cardiac tumors—28 patients with CMs and 27 patients with CPFEs (1, 11–13). Similar to prior investigations, most patients in this study presented with embolic phenomena (54.5%) (11, 14–17). Patients were in their fifth decade of life, and about two-thirds were women. We found NLR to correlate with tumor size for both CM and CPFE and predict CVE recurrence. To the best of our knowledge, these are novel findings. Our study showed that patients with CPFEs were more likely to have valvular dysfunction (moderate or severe valve stenosis or regurgitation as assessed in accordance with established guidelines) or a history of cardiac surgery (9, 10). This may lend credence to the theory that endothelial and valvular damage promotes a nidus for microthrombi collection that raises the risk for CPFE. Two reports have also associated CPFEs and β-thalassemia blood disorders, suggesting a similar hemodynamic mechanism (18, 19). However, a recent study investigating the molecular profile of 14 CPFE patients demonstrated mutations of the KRAS oncogene in 11 patients, supporting the neoplastic nature of these lesions (20). Based on echocardiographic reports in this study, CMs were often left-sided (85.7%) and most commonly in the left atrium (71.4%). CPFEs were exclusively left-sided, predominantly valvular (92.6%), and most commonly on the aortic valve (70.4%). These findings correlate with the existing literature (4, 11, 17, 21–23).

No specific biomarkers for benign or malignant cardiac neoplasms are currently validated. Leukocytosis is a nonspecific indicator of inflammation. NLR is a readily accessible adjunct prognostic biomarker of growing contemporary interest. It reflects a shift from adaptive to innate immunity due to a combination of disordered immunity and contributions from disease-specific etiologies. The normal range of NLR is typically between 1 and 2, while values higher than 3.0 and lower than 0.7 are considered pathological. Elevated NLR has been demonstrated as a prognostic marker for morbidity and mortality in various chronic diseases (e.g., atherosclerotic disease, cerebrovascular disease) and cancers (e.g., melanoma, breast cancer) (24). In the case of malignancy, neutrophilia is an observed feature of cancer-elicited chronic inflammation, and it is often accompanied by relative lymphocytopenia due to tumor-driven suppression of the cell-mediated immune response (25, 26). NLR has been associated with tumor size, tumor stage, and metastatic potential and has been shown to predict cancer-specific survival, including progression-free and disease-free survival (27).

In this study, the white blood cell count and NLR were elevated in patients presenting with CMs and CPFEs, although values were significantly higher in the former. This is likely due to the substantially larger tumor sizes of CMs than those of CPFEs. It may also naturally reflect the immunologic nature of CMs, which has been described to overproduce interleukin-6 and other various growth factors and cytokines that generate observed constitutional symptoms. Indeed, interleukin-6 has been posited as a marker of recurrence for CMs (28). In CPFEs, the elevated NLR may be a product of endothelial cell activation and chronic valvular inflammation in addition to neoplastic processes (29). Elevated NLR has previously been observed in patients with CMs, and it has been correlated with tumor size (30). NLR values were similarly correlated with various metrics of tumor size in this study (64%–67% in the CM and 56%–59% in the CPFE group), suggesting a role for this biomarker in tumor burden prognostication. Of note, NLR values may also be elevated in ischemic chest pain, CVE, and other stress states present in these patients during hospitalization; however, NLR values were not significantly altered when stratified by the patient presentation (31). Although CMs and CPFEs generally differ in fundamental pathophysiology, there is considerable overlap in the initial clinical manifestation. Moreover, unusual presentations of these tumors (e.g., the deranged echocardiographic appearance of CMs or nonvalvular site of CPFEs) obscure identification (32, 33). Therefore, the patterns of NLR elevation observed in this study may be useful in the setting of a patient in whom a primary cardiac neoplasm is detected, but the exact etiology is unclear.

Importantly, a sizeable portion of patients with either neoplasm (30%–60%) initially present with CVEs, consistent with the results of this investigation. CM-related CVEs have been associated with a friable or villous tumor surface, younger male patients, and tumors with size ≤4.5 cm (14, 15, 34). CPFE-related CVEs have been associated with small tumor size at diagnosis and independent mobility (11, 17). Other echocardiographic characteristics, aortic valve location, or tumor growth rate have not been identified as risk factors (35). In this study, tumors were expectedly left-sided. No baseline patient characteristics, including age, sex, or cardiovascular risk factors such as a history of hypertension, hyperlipidemia, coronary artery disease, diabetes mellitus, smoking, or prior CVE, were significantly associated with an initial cerebrovascular manifestation of the benign cardiac neoplasm compared to any other indication for presentation. The size or independent mobility of the tumor and left atrial enlargement also did not significantly differ between patients initially presenting with CVEs and any other indications.

Elevated recurrence of CVE, compared to baseline age- and sex-matched controls, after excision has also been reported in both CMs and CPFEs. Prolonged duration until surgical excision in patients with CMs and unremoved lesions in patients with CPFEs are associated with CVE recurrence (4, 11, 16). The CVE recurrence rate in our cohort was high: 30.0% at 5 years (24.0% in patients with prior tumor resection and 60.0% in patients without prior resection). The prolonged duration between symptom onset and surgical resection (>60 days), left atrial enlargement, male sex, and NLR > 3.0 at approximately 1-year follow-up were significantly associated with CVE recurrence within 5 years of surgical resection or discharge. When controlling for age and known risk factors for CVE in multivariable analysis, left atrial enlargement and NLR > 3.0 at approximately 1-year follow-up remained significantly associated with 5-year CVE recurrence. Ultimately, these findings may emphasize the need for timely excision of CMs and CPFEs in suitable candidates and the use of NLR as a prognostic biomarker. Importantly, no cases of hemorrhagic transformation intraoperatively or postoperatively have been reported, lessening concern for postischemic stroke patients on bridge therapy.

This investigation has several limitations that must be emphasized. The study design is retrospective and small in sample size, limiting the robustness of the inferences that can be drawn. Our tertiary care center may also be subject to referral bias, possibly explaining the relatively low rates of incidental discovery of these benign cardiac neoplasms compared to other studies. Regarding NLR, the prognostic biomarker is an accessible and cost-effective measure of dysregulated inflammation that has garnered significant contemporary interest in various fields. In numerous large-cohort studies and meta-analyses within the cardiovascular domain, NLR has been significantly associated with the onset of atrial fibrillation, acute coronary syndrome (including correlation with the Global Registry of Acute Coronary Events risk score and SYNTAX score), and acute decompensated heart failure (36–40). However, its use in cardiovascular tumors has not been well characterized, and further large-scale investigations are required to validate the observations within our study. The reference cutoff and the subsequent interpretive potential of NLR may also vary by age, sex, and other factors, although this requires further research (41).

Furthermore, NLR may be elevated in various acute disease states; therefore, its utility may primarily be as an adjunct prognostic tool (after further validation) and not in a diagnostic capacity. A presentation of ischemic stroke is among the acute etiologies associated with elevated NLR (42). Therefore, it would be of great interest to evaluate NLR in patients presenting with CVEs and underlying benign cardiac neoplasms compared to values in patients with CVEs and no malignancy to further validate this work. Finally, this study included a small number of patients in whom the benign cardiac neoplasm was not excised. In these cases, the diagnosis was formulated via clinical presentation and the presence of characteristic echocardiographic features of the tumor (8). A few contemporary studies have discussed the viability of a conservative approach to asymptomatic benign cardiac neoplasms (especially small, incidentally discovered tumors), one that favors antithrombotic medical therapy as opposed to surgical intervention (16, 17, 43, 44). While our investigation failed to highlight a significant elevation in adverse outcomes in patients in whom surgical excision was not pursued, likely due to the small sample size, we feel it is important to include these cases to reflect these considerations; we bear in mind that without histopathological confirmation of tumors in these cases, the uncertainty and risk of misdiagnosis remains.

In conclusion, our study demonstrated a high CVE recurrence rate in patients with benign cardiac tumors and indicates that NLR may serve as a valid adjunct prognostic tool for tumor burden and recurrence of neurologic events.