Cancer and cardiovascular diseases stand as leading causes of mortality. Advances in medical technology have notably reduced cancer-related deaths through chemotherapy, enhancing patient outcomes—a pivotal aspect of effective tumor care [1]. Nevertheless, chemotherapy drugs can harm various organs, with cardiotoxicity being a notable concern [2]. Detecting this early is crucial for both oncologists and cardiologists.

Given that chemotherapy’s cardiotoxic effects span the myocardium, blood vessels, cardiac rhythm, and blood pressure, evolving cardiovascular imaging methods aid in spotting and managing such cardiotoxicity. Techniques like Single-Photon Emission Computed Tomography (SPECT) and positron emission tomography (PET) offer significant value in spotting abnormal heart perfusion and evaluating ischemic risks [3]. PET molecular imaging, pivotal in oncology, not only aids in diagnosis but also directs treatment and monitors its efficacy. The widely used tracer in PET/CT applications, Fluorine-18 Fluorodeoxyglucose (¹⁸F-FDG), distinguishes tumor tissues based on distinct metabolic activity from surrounding areas. In cardiovascular contexts, it gauges inflammation and ischemia by assessing heightened metabolic activity and glucose uptake in the heart and vessels [4]. While preparation methods differ between tumor and heart imaging, a prolonged fast before tumor scans minimizes skeletal and heart muscle glucose uptake, leaving ischemia and inflammation-related uptake unaffected [5, 6]. Despite guidelines emphasizing the role of cardiac nuclear imaging in recognizing and managing tumor-related heart issues during treatment [7], not all cancer patients undergo standardized cardiac nuclear tests. Hence, analyzing abnormal cardiac uptake in Fluorine-18 Fluorodeoxyglucose Positron Emission Tomography/Computed Tomography (¹⁸F-FDG PET/CT) scans of tumor patients remains a significant area for exploration.

In ¹⁸F-FDG PET/CT scans of tumor patients, we noted varying patterns and intensities of abnormal cardiac ¹⁸F-FDG uptake both pre- and post-radiotherapy and chemotherapy. Some cases of abnormal uptake align with existing heart conditions, while others lack a clear cause. Understanding these uptake patterns is crucial to avoid misdiagnoses and lay the groundwork for effectively using ¹⁸F-FDG PET/CT to pinpoint cardiotoxicity related to tumors or cancer treatments. Currently, research on abnormal cardiac uptake prior to anticancer interventions in standard oncologic ¹⁸F-FDG PET/CT scans is sparse. Hence, leveraging imaging and clinical data, this study seeks to delineate the features and clinical relevance of abnormal cardiac uptake preceding anticancer treatments in oncologic ¹⁸F-FDG PET/CT evaluations.

Anti-cancer medications can induce cardiotoxicity, manifesting as myocardial inflammation, irreversible damage, and subsequent cardiac impairment [16]. Treatment-related vasoconstriction, along with arterial and venous thrombosis, can precipitate coronary artery disease and acute myocardial infarction [17]. Moreover, treatment effects on blood pressure and the cardiac conduction system may lead to hypertensive crises and severe arrhythmias [18]. Consequently, the European Society of Cardiology guidelines advocate for cardiovascular risk assessment in cancer patients and recommend cardiovascular imaging before and after treatment [19]. Beyond echocardiography, electrocardiography, and cardiac magnetic resonance imaging, nuclear cardiac imaging serves as a prevalent diagnostic tool for assessing drug-induced cardiotoxicity.

While ¹⁸F-FDG PET/CT imaging primarily targets malignant tumors for staging, assessing burden, and gauging treatment response, its utility has expanded to evaluating cardiotoxic effects of anthracyclines and other anti-cancer agents. These drugs can disrupt mitochondrial oxidative stress metabolism and elevate glycolytic activity in myocardial cells [20]. Given that hexokinase, the rate-limiting enzyme of glycolysis, governs both ¹⁸F-FDG phosphorylation and retention within cells, heightened myocardial ¹⁸F-FDG uptake can indicate drug-induced cardiotoxicity [21]. Research by Bauckneht et al. [22] demonstrated dose-dependent increases in left ventricular ¹⁸F-FDG uptake in mice administered varying doxorubicin concentrations. Similarly, a study involving 36 Hodgkin lymphoma patients undergoing ABVD treatment and sequential ¹⁸F-FDG PET/CT scans revealed elevated left ventricular SUV during doxorubicin therapy, correlating with subsequent cardiac events [22]. Bauckneht et al. [22] further noted that lower baseline ¹⁸F-FDG uptake corresponded with rising doxorubicin dosages, suggesting baseline uptake could influence chemotherapy-linked cardiac toxicity. Additionally, Fathala et al. [23] identified a significant association between myocardial ¹⁸F-FDG uptake patterns—focal or diffuse—and perfusion imaging-based myocardial ischemia, indicating that baseline and treatment-related ¹⁸F-FDG uptake patterns crucially inform tumor-related cardiac toxicity. However, comprehensive studies analyzing heart uptake patterns before initiating anti-cancer treatment remain limited.

Heart uptake patterns can be categorized as no uptake, diffuse uptake, focal uptake, and focal uptake on a diffuse background. Both no uptake and diffuse uptake are typically deemed physiological. However, specific instances of focal uptake also fall under this category. Notably, uptake patterns remain unaffected by factors like age, blood glucose levels, body weight, or FDG dosage, resulting in poor repeatability in PET/CT examinations [24]. Although focal FDG accumulation in certain regions like the left papillary muscle’s anterior lateral and posterior inferior areas is considered physiological, isolated FDG uptake in this muscle could indicate a tumor or thrombus [25].

The crista terminalis, a smooth muscular band separating the right atrium from its appendage, may occasionally show heightened FDG uptake. This can lead to misinterpretations as myocardial tumors, thrombi, or focal pericardial metastases, despite its physiological nature [26]. Additionally, lipomatous hypertrophy of the interatrial septum (LHIS), a benign condition with a prevalence of around 1–8%, presents challenges due to its appearance as a dumbbell-shaped thickening of the interatrial septum. It displays focal FDG uptake linked to brown adipose tissue, potentially causing confusion with other conditions [27]. Furthermore, basal myocardium regions often manifest circular, semicircular, or spotted FDG uptake, which is another recognized physiological pattern [24].

Given these considerations, this study categorizes cardiac FDG uptake patterns as either physiological or abnormal. Abnormal cardiac uptake may occur in the left ventricle, right ventricle, or atrium. Analyzing patients’ medical histories, cardiac ultrasounds, coronary angiographies, and other imaging studies helped identify prevalent causes of abnormal uptake in various cardiac regions. Specifically, abnormal uptake in the left ventricle commonly stems from myocardial ischemia, hypertensive ventricular hypertrophy, and heart failure. In contrast, abnormal uptake in the right ventricle often relates to conditions like pulmonary arterial hypertension and valvular diseases. Common causes of abnormal uptake in the atrium encompass atrial fibrillation, hypertensive atrial enlargement, and pulmonary arterial hypertension. Elevated pressure or volume loads on the atrium or ventricle trigger the synthesis and upregulation of glucose transporters, culminating in heightened myocardial cell metabolism [28]. In scenarios involving ventricular hypertrophy, augmented myocardial workload can further amplify myocardial cell metabolism. Additionally, increased FDG uptake observed in atrial fibrillation patients may correlate with enhanced epicardial adipose tissue activity [29].

Upon analyzing abnormal cardiac uptake causes among tumor and non-tumor patients, it became evident that tumor patients exhibit a significantly higher rate of unexplained cardiac uptake compared to their non-tumor counterparts. Given this heightened abnormal cardiac uptake in tumor patients, a plausible link to tumor burden emerges. Further stratifying tumor patients by tumor burden using ¹⁸F-FDG PET/CT revealed that the high TLG group manifested a notably elevated rate of unexplained cardiac uptake relative to the low TLG group. Consequently, unexplained cardiac uptake in tumor patients undergoing ¹⁸F-FDG PET/CT likely correlates with tumor burden, suggesting that increased tumor burden elevates the incidence of abnormal cardiac uptake [30]. Unexplained cardiac uptake may be the manifestation of tumor-associated myocardial metabolic remodeling, which essentially reflects a dual process of adaptive compensation and early injury of the myocardium under the stimulation of the tumor microenvironment or damage [31]. Tumors sequester peripheral glucose through the Warburg effect while inhibiting myocardial fatty acid oxidation. To compensate for the energy deficit, the myocardium upregulates glucose transporter 4 (GLUT4) expression and activates the PI3K/Akt pathway to enhance glucose uptake and glycolysis. In this context, increased glucose uptake represents a compensatory adaptation of the myocardium to energy substrate imbalance. Reactive oxygen species (ROS) released by tumors cause mild damage to myocardial mitochondria, leading to reduced oxidative phosphorylation efficiency [32]. Consequently, the myocardium actively switches to glycolysis for energy production—even under oxygen-sufficient conditions. Here, increased glucose uptake serves as a metabolic “risk-aversion” strategy after mitochondrial dysfunction, preventing energy crisis triggered by reliance on the defective oxidative phosphorylation pathway. While these findings align with prior research, the precise underlying mechanism remains elusive.

In-depth analysis comparing unknown cause abnormal cardiac uptake to known cause abnormal cardiac uptake in tumor patients revealed no notable differences in metabolic parameters such as SUVmax and SUVmean between the groups. SUV merely indicates individual pixel metabolic levels within images, lacking the capacity to encapsulate the intricate details or overall status of abnormal cardiac uptake lesions. Thus, solely gauging changes in SUV values falls short in distinguishing between known and unknown causes of abnormal cardiac uptake.

Conversely, texture analysis unveils nuanced image information imperceptible to the unaided eye by delineating relationships among voxels and pixels concerning grayscale, intensity, and spatial positioning. This analytical approach equips clinicians with the tools to discern subtle disparities among akin lesions. Despite limited research on texture analysis of abnormal cardiac uptake lesions, this study determined that integrating Teta3, 45dgr_RLNonUni, 45dgr_GLevNonU, S(1,-1)Entropy, SUVmean, and heart-to-blood pool SUV ratio within a predictive model yielded optimal discriminatory outcomes, boasting an AUC of 0.809 alongside elevated sensitivity and specificity. This underscores the efficacy of combining texture features with metabolic parameters in ¹⁸F-FDG PET/CT for pinpointing abnormal cardiac uptake causes.

Notably, uniformity metrics from the run length matrix and gray level matrix in texture analysis can signify lesion pixel consistency, with the known cause abnormal uptake group exhibiting markedly greater non-uniformity than its unknown cause counterpart. Elevated values indicate increased image irregularity. Furthermore, when assessing lesion metabolic parameters, a heightened ratio of lesion SUVmax to corresponding blood pool SUVmax augments the probability of identifying known cause abnormal uptake. This comprehensive approach emphasizes that amalgamating texture with metabolic parameters enhances predictive accuracy for abnormal cardiac uptake causes. Crucially, this method may provide some assistance in early identification of the underlying causes of abnormal cardiac uptake in patients during follow-up ¹⁸F-FDG PET/CT, before other imaging evidence of organic damage emerges.

This study presents several limitations worth noting. Firstly, being a retrospective clinical study, it lacks the capability to delve into the pathological or histological dimensions of abnormal cardiac uptake of ¹⁸F-FDG in patients, thereby hindering a clear understanding of its underlying mechanisms. Secondly, despite incorporating an extended fasting period ($\ge$12 hours) in the imaging protocol, complete suppression of physiological uptake of ¹⁸F-FDG by the left ventricular myocardium may not have been achieved. Thirdly, simultaneous myocardial fatty acid imaging was absent from the study.

While ¹⁸F-FDG PET/CT imaging remains pivotal for diagnosing, staging, and assessing treatment responses in tumors, its role extends to evaluating cardiac abnormalities manifesting through abnormal ¹⁸F-FDG uptake. Although both unknown cause and known cause abnormal cardiac uptakes may correlate with tumor burden, the latter is more likely associated with conditions like myocardial ischemia, hypertension, and atrial fibrillation. Distinguishing between these two necessitates more than just metabolic parameters; it demands a comprehensive assessment integrating both metabolic parameters and texture features. Such an approach furnishes a theoretical framework for elucidating the emergence of new cardiac abnormal uptakes and alterations in cardiac abnormal uptake patterns during tumor treatments.

This study analyzed the possible causes and characteristics of myocardial abnormal uptake in tumor patients during their initial ¹⁸F-FDG PET/CT imaging, providing theoretical support for understanding changes in cardiac abnormal uptake during tumor treatment. It is expected to offer additional value in observing tumor-related cardiac toxicity through ¹⁸F-FDG PET/CT imaging.

The datasets generated and analysed during the current study are available from the corresponding author upon reasonable request.

RLL, AJT, JY and WLX designed the research study. JW and YWB conducted the research, while AJT and JY provided guidance and advice. RLL and WLX analyzed the data. YWB, JW drafted the manuscript. All authors contributed to the critical revision of the manuscript for important intellectual content. All authors have read and approved the final manuscript. Furthermore, all authors have contributed sufficiently to the work and have agreed to be accountable for all aspects of the research.

This study received approval from the Ethics Committee of the Second Hospital of Dalian Medical University (No.2019-049), and all individual participants included in the study provided signed informed consent. The study adhered to the principles outlined in the Declaration of Helsinki.

Thanks to all the peer reviewers for their opinions and suggestions.

This work was supported by grants from the Natural Science Foundation of Liaoning Province (2024-BS-184), Basic Scientific Research Program of the Education Department of Liaoning Province (No. LJKMZ20221288), and “1+X” Plan of the Second Hospital of Dalian Medical University (2024LCJSYL16).

The authors declare no conflicts of interest.