Myocardial infarction with non-obstructive coronary arteries (MINOCA) is recognized if it meets the general criteria of myocardial infarction (MI) together with the absence of significant lesions in epicardial arteries in angiography (1). As shown in large MI registries, MINOCA concerns 1–13% of all patients with MI (2, 3). Recent reports indicate an unexpectedly unfavorable long-term prognosis in this group of patients. The SWEDEHEART registry included 9,092 patients with MINOCA, of whom 24% experienced a major cardiovascular event, and where 14% died within a mean follow-up period of 4.5 years (4).

The potential mechanisms responsible for MINOCA are heterogeneous (1, 5). According to the current knowledge, the underlying pathophysiological causes of MINOCA are grouped as coronary or non-coronary. Moreover, the latter are classified as myocardial disorders or as those that are typically extra-cardiac (6). Both historical (7) as well as current findings (8) indicate that hypercoagulable states, including the inherited thrombophilia, occurred in 15–25% of patients with MINOCA. This includes deficiency of protein C, protein S, or antithrombin. Additionally, the antiphospholipid syndrome was detected in 15.5% of patients. Concurrently, patients with cancer are a group that is at a particularly high prothrombotic risk, traditionally in the venous system (9). An analysis of the Surveillance, Epidemiology, and End Results involving nearly 280,000 patient pairs showed that the rate of arterial thromboembolic events was 4.7% in cancer patients compared with the 2.2% in controls (10). That predisposition for arterial thromboembolism, defined as MI, ischemic stroke, or peripheral arterial occlusion, has been confirmed recently in a large Danish population-based cohort study (1.5 vs. 0.8% in the 6-month observation, hazard ratio [HR]: 2.36, 95% confidence interval [CI]: 2.28–2.44] (11). Moreover, its occurrence among patients with cancer was associated with an increased risk of mortality (HR 3.28, 95% CI: 3.18–3.38) (11). As the arterial thromboembolic events immediately preceded cancer diagnosis and were correlated with the stage of cancer (10, 11), they can be considered paraneoplastic symptoms, which always require subsequent meticulous diagnostics toward a subclinical neoplastic process (12).

Recently, a review of the meta-regression analysis of nine studies including 26,636 patients with MINOCA has shown that 2.5% of them had a diagnosis of malignancy at presentation (13). Similar findings have been reported in the SWEDEHEART registry (14). Despite relatively low prevalence, both Nordenskjöld et al. (4) (HR: 2.40, 95% CI: 1.58–3.61, p < 0.001) and Pelliccia et al. (13) (coefficient: 0.001, 95% CI: −0.001 to 0.001, p = 0.01) have found cancer as an independent predictor of death in patients with MINOCA. Another meta-analysis including a higher number of patients with MINOCA, i.e., 36,932, did not confirm a similar relationship (15). Therefore, we sought to characterize subjects with MINOCA and cancer hospitalized in a tertiary cardio-oncology center in order to investigate the potential mechanisms affecting their long-term outcomes.

As has been stated retrospectively, in a tertiary cardio-oncology center including closely cooperating departments of cardiology (168 hospital beds), cardiac surgery (80 beds), pulmonology and oncology (74 beds), and thoracic surgery (48 beds), 1,011 consecutive patients underwent coronary angiography between 2012 and 2017 due to the diagnosis of MI based on clinical symptoms, electrocardiographic findings, and the evolution of myocardial necrotic biomarkers (16). MINOCA was recognized in 72 (7.1%) subjects (Figure 1) based on the universal criteria of MI (positive cardiac biomarkers rising and/or falling in serial measurements, with at least one value above the 99th percentile as the upper reference limit and at least one clinical sign of infarction). An additional inclusion criterion was a lack of obstructive lesions narrowing epicardial coronary segments by more than 50% in angiography (1, 17). Patients with ST-segment elevation of at least 1 mm in at least two contiguous leads were classified as ST-segment elevation MI (STEMI), whereas patients without ST-segment elevation on admission were diagnosed as non-ST-segment elevation MI (NSTEMI) (18). In addition, 134 (13.3%) patients were identified with active cancer, defined as cancer diagnosed within the past 6 months, receiving antimitotic treatment during the last 6 months, recurrent, metastatic, regionally advanced, or inoperable (19) (Figure 1). In the analyzed period of time, five MI patients with advanced cancer did not undergo coronary angiography and were therefore excluded from further analysis. The study protocol complied with the Declaration of Helsinki and was approved by the Jagiellonian University Medical College Ethics Committee (Consent No. 1072.6120.59.2018). All included patients gave informed consent.

Based on detailed angiographic and oncological characteristics, four groups of patients were created (Figure 1). Within 1,011 MI patients, active cancer and MINOCA were identified in 21 (2.1%) patients, whereas MINOCA without cancer was diagnosed in 51 (5.0%) subjects. Of the 939 remainders with type 1 MI with obstructive coronary artery disease (MI-CAD), 113 (11.2%) patients had active cancer and 826 (81.7%) had no evidence of active cancer. In 111 patients, the malignancy process was diagnosed before index MI, whereas new cancer was found during index hospitalization in two patients with MINOCA and in 21 with MI-CAD (Table 1).

Among the four groups, there were significant differences in the distribution of gender, anthropometric parameters, dyslipidemia, active smoking status, and initial clinical presentation (p < 0.01 for each) (Table 1). The angiographic analysis also revealed a different proportion of epicardial thrombus in the compared groups (p = 0.02). Hemoglobin levels were lower, whereas baseline high-sensitive troponin T was higher in both cancer groups compared with non-cancer MINOCA subjects (p < 0.05 for all pairwise comparisons) with the blurring of differences during hospitalization in maximal peak values (Table 2). After adjustment for renal function, the highest inverse correlation between hemoglobin level and baseline troponin T concentration was found in the cancer MINOCA (r = −0.41, p = 0.05) group (Figures 2A–D). The proposed ratio of troponin T to hemoglobin was higher in cancer patients with MINOCA and MI-CAD when compared with the respective non-cancer groups (Figure 2E). The time of hospitalization was insignificantly shorter in non-cancer MINOCA (4 (3–7) days) as compared with cancer MINOCA (6 (3–12) days), cancer MI-CAD (6 (3–9) days), and non-cancer MI-CAD (6 (4–8) days) and (p = 0.07).

To our knowledge, this study is the first and most comprehensive analysis derived from a tertiary cardio-oncology center concerning the complex relationship between cancer and MINOCA, as well as its influence on long-term clinical outcomes. As shown, neoplasm has been identified more frequently in patients with MINOCA than in those with atherosclerosis and/or thrombus-based type 1 MI (defined as MI-CAD). However, a multivariable analysis showed that an active malignancy was associated with unfavorable long-term outcomes. We have also provided clinical features that characterized cancer patients with MINOCA, which might be useful in their differential diagnosis. It is important to note that the diagnosis of cancer in both MINOCA and MI-CAD groups was associated with an extremely high all-cause mortality in a 5-year observation. Moreover, a multivariable approach limited to only the MINOCA group showed that active cancer irrespective of age and lower left ventricular systolic function affected a higher mortality rate.

Patients with MI-CAD and cancer distinguished in our study were characterized by a highly unfavorable prognosis driven mostly by neoplastic disease. Although treatment of such patients should be strictly individualized, there are still limited data sufficiently addressing the optimal management of MI in patients with cancer (27). Further studies are warranted to establish an optimal antithrombotic regimen, especially in the acute phase, due to the proven high risk of stent thrombosis (9, 28). The results derived from the large Nationwide Inpatient Sample indicate that cancer in patients receiving percutaneous coronary intervention is common, but its prognostic impact depends on detailed oncological characteristics (29). Our results also indicate that, in both cancer and non-cancer MI-CAD patients, the rate of revascularization with the percutaneous coronary intervention was almost 90% emphasizing current trends in interventional cardiology. While cancer patients with type 1 MI were historically less likely to receive primary percutaneous coronary intervention with first-generation drug-eluting stents mainly due to the need for a shorter course of dual antiplatelet therapy following bare-metal stents, the new drug-eluting stents requiring shorter antiplatelet therapy time have become more effective and as safe as bare-metal stents. According to the current registries, dual antiplatelet therapy was prescribed in only half of the patients with MINOCA, mainly in those with sinus rhythm, prior percutaneous coronary intervention, and active smokers (30).

In contrast, the prognosis in patients with MINOCA remains controversial, with the latest studies suggesting comparable (4, 31) or lower (15) long-term mortality rates in patients with MINOCA vs. MI-CAD. The abovementioned studies indicate that a history of cancer coexisting with 2–2.5% of patients with MINOCA (4, 13) is (13) or is not (15) an independent predictor of their long-term mortality. In our MINOCA and MI-CAD groups, a diagnosis of active cancer made before index MI was more common. This overrepresentation of neoplastic status was independently associated with unfavorable long-term survival. When compared with the available literature, such a high proportion of cancer patients is primarily a result of the structure of our center, as well as that of direct admissions from oncology and thoracic surgery departments to the cardiology ward. Interestingly, there is a visible trend toward more frequent admissions of cardio-oncology patients due to their prolonged survival time.

The etiology of MI in the oncological population is multifactorial. In previous studies, the role of cancer-induced immunological disorders, oxidative stress, prothrombotic state, and oncological treatment was underlined in MI development among cancer patients (32). Moreover, oncological patients are generally high-risk due to the significant prevalence of traditional cardiovascular risk factors, such as older age, hypertension, dyslipidemia, diabetes, obesity, or tobacco addiction (28). This was also corroborated in this current study. Most of the above-indicated factors contribute to the shifted oxidase-reductase balance and endothelial injury. This exacerbates coronary artery disease progression and promotes the rupture of atherosclerotic plaque associated with type I of MI, identified as MI-CAD (28, 32, 33) in our study. On the contrary, the influence of cancer and antitumor treatment is undeniable among MINOCA survivors. The rupture of non-obstructive plaque, distal embolization, hypercoagulable state with thrombus formation, transient artery spasm, microvascular dysfunction often caused by endothelial impairment, and supply-demand mismatch, among others, are all mechanisms responsible for MINOCA (5). It is worth noting that, each of these sequences of events might be triggered by both tumor and antineoplastic treatment (13). The classic chemotherapy drugs have been proven to damage the coronary arteries, mainly in their endothelium. Therefore, they can lead to acute thrombosis and coronary spasms (33). Drugs that particularly increase the risk of MI include fluoropyrimidines (5-fluorouracil, capecitabine, gemcitabine) and platinum compounds (33), which were also often used among the analyzed patients. Moreover, combining chemotherapeutics from different groups, especially those mentioned above, significantly increases the risk of MI (33). However, there is a lack of original reports demonstrating the relationship between chemotherapy and MINOCA. Our study provides detailed angiographic and echocardiographic characteristics of cancer patients with MINOCA, shedding light on their potential relationships. These findings might be helpful in further research dedicated for personalized treatment in this demanding group of patients.

A long-term prognosis is associated with myocardial infarct size. As we have shown, both cancer and non-cancer patients with MINOCA were characterized by a better preserved global LV function and lower peak high-sensitive troponin levels compared with the corresponding MI-CAD groups. This indirectly indicates a lower myocardial injury rate and most likely a smaller infarct size in patients with MINOCA. These findings are in line with previous data showing that, among the MINOCA population, patients with heart failure with preserved LV ejection fraction (34, 35) predominated. Post-infarction myocardial remodeling is also less frequently observed in this group. There are at least a few potential explanations for this relationship. First, the smaller myocardial infarct size is a consequence of a higher prevalence of NSTEMI in MINOCA (8). Second, cardiac magnetic resonance imaging provides evidence that, in patients with MINOCA, only small foci of necrosis are often observed, while myocardial edema is the dominant abnormality (36).

In this study, hemoglobin levels were lower in both cancer groups, compared with respective non-cancer MINOCA and MI-CAD groups. Moreover, as has been shown in our multivariable models, lower hemoglobin levels worsen long-term prognosis in the whole group, but not in the population limited to patients with MINOCA. According to criteria similar to ours, anemia at baseline was found in approximately 40% of patients in the European Cancer Anemia Survey (37). This proportion increased up to 60–70% during either anticancer treatment or cancer progression, affecting the higher overall mortality risk (38). In our cancer patients with MINOCA, lower hemoglobin levels were associated with higher baseline troponin concentrations, suggesting the possibility of anemia-induced myocardial injury (16). As has been shown previously, active cancer should be considered as a secondary cause of troponinosis that is not associated with acute coronary syndrome (39, 40). Moreover, troponin elevation was linked with a higher mortality rate, especially in patients with lung cancer (41). We have also found that the ratio of troponin T to hemoglobin was significantly higher in both cancer populations when compared to the respective non-cancer groups. Our findings are one more argument for the adoption of a higher troponin cut-off value for MI in patients with cancer (39, 40). The relatively high proportion of patients with cancer-induced anemia, also visible in our cohort, may require blood transfusion or other available methods of treatment (erythropoietin or iron supplementation). In a propensity-matched analysis, Salisbury et al. have demonstrated that blood transfusion was associated with a lower risk of in-hospital mortality (42). In turn, a meta analysis done by Chatterjee et al. indicates that a liberal blood transfusion strategy is associated with higher all-cause mortality when compared to a more restricted strategy, which might be associated with volume overload, increased thrombogenicity, impaired oxygen delivery, and a risk of infection (43).

Our findings provide evidence that active cancer in the whole cohort of patients with MI, overrepresented among the MINOCA population, is associated with extremely high long-term mortality. A multivariable approach indicates that an active malignancy was independently associated with unfavorable long-term survival in the whole MI population as well as in patients with MINOCA.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by Jagiellonian University Medical College Ethics Committee (Consent No. 1072.6120.59.2018). The patients/participants provided their written informed consent to participate in this study.

KS conceived the concept of the study. KS, KN, and JZ contributed to the design of the research. KS, KN, and BS reviewed the literature and were involved in data acquisition. All authors analyzed and interpreted the data. JN and JZ supervised data processing. All authors edited and approved the final version of the manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.