Editor's response to letter

Thank you for your letter regarding the study by Tolstrup et al., “Accurate diagnosis of ischemic heart disease without exposure to radiation using non-stress unshielded magnetocardiography (MCG).” We appreciate your interest in this innovative research and your thoughtful observations. The authors' work indeed represents a significant step forward in non-invasive cardiac diagnostics, offering thought provoking and future direction for a potentially safer and more accessible and efficient alternative to current methods. We welcome the opportunity to address the points you have raised, as constructive feedback is crucial for advancing scientific knowledge and refining novel techniques. In the following response, we will address each of your observations in turn, providing clarification and additional context where necessary, and discussing how these insights can inform future research in this promising field.

The study includes patients both with and without ischemic heart disease, consisting of 99 patients presenting with acute chest pain, evaluated at a large hospital in the US using a 9-channel MCG device, and 34 patients with stable angina, studied at a university hospital in Italy using a 36-channel device. The distinct patient populations, acute chest pain in the US versus stable angina in Italy, may introduce regional or clinical differences, further limiting the ability to generalize the findings across diverse patient groups or settings.

We appreciate the insightful observation regarding potential regional and clinical differences in the study population stemming from the distinct patient cohorts included in this research.

One of the aims of the study was to show that the predictive accuracy of MCG is comparable with that of myocardial perfusion scans in a standard clinical practice to detect ischemia. To achieve this, a heterogeneous sample of patients with cardiac ischemia and non-ischemic conditions was necessary. The inclusion of patients from different clinical settings—acute chest pain cases and stable angina cases was a deliberate aspect of the study design intended to reflect the diverse clinical presentations encountered in real-world practice.

Most patients presenting with acute chest pain to US Emergency Departments do not ultimately have acute coronary syndrome. Instead, they often receive diagnoses such as stable angina, non-ischemic cardiac conditions (e.g., pericarditis, myocarditis, arrhythmias), or non-cardiac etiologies. While this distribution aligns with our goal of evaluating MCG in a representative population, we acknowledge and agree that a more balanced representation of acute and stable cases across both study sites could have further strengthened the generalizability of our findings.

Another limitation of the study is related to the group of healthy controls. Obviously, confirmatory tests were not performed to exclude significant coronary artery disease (CAD) in presumed healthy controls as they were assumed free of disease. If this assumption is incorrect, the negative predictive value (NPV) of the MCG system could be overestimated. In an ideal world, controls should have been validated through negative functional testing to strengthen the conclusions, but an alternative could be to have a higher sample size in which case cohort with non-ischemic chest pain could serve this purpose.

We appreciate this insightful observation regarding the healthy control group in the study. The authors acknowledge that the definition of “healthy controls” used was indeed pragmatic, reflecting a clinically reasonable approach that aligns with standard clinical practice. The 63 individuals classified as healthy controls were non-smoking, asymptomatic, and had no known hypertension, diabetes, or heart disease. While we recognize the possibility that some degree of coronary ischemia might exist within this cohort, the likelihood is very low and exceeds the typical clinical standard of care for defining healthy individuals.

To address this potential limitation, the authors calculated the Negative Predictive Value (NPV) both with and without the inclusion of healthy controls. As reported in section 3.4 of the study, the NPV without healthy subjects was 89 %, while including them yielded an NPV of 94 %. This transparent approach allows readers to assess the potential impact of the healthy control group on the study's conclusions.

While the impact on NPV is likely not substantial, we acknowledge that some effect may indeed be present. This underscores the need for larger studies with improved randomization and more definitive reference standards. Such future research would not only address the limitations identified but also provide a more comprehensive evaluation of MCG's diagnostic capabilities in diverse clinical settings.

Common arrhythmias, such as ST-elevation myocardial infarction (MI), atrial flutter, left bundle branch block (LBBB), and third-degree atrioventricular block, were excluded from the study without any explanation. If the authors are aware of evidence suggesting that these conditions interfere with MCG findings, this should have been clarified. Otherwise, excluding these arrhythmias limits the clinical relevance of the results, as they are frequently encountered when evaluating ischemic chest pain.

Thank you for recognizing this. It's important to acknowledge that while studies have shown that MCG can be used to diagnose cardiac arrhythmias the focus of this study was on detecting coronary ischemia [1,2].

The exclusion of hemodynamically unstable patients, those with ST-elevation myocardial infarction (STEMI), and third-degree atrioventricular (AV) block is justified, as these conditions represent medical emergencies requiring immediate intervention. This approach aligns with ethical clinical practice and ensures patient safety. Of these, only third-degree AV block is an arrhythmia, but all of them represent emergencies that need immediate intervention.

Within clinical trials evaluating MCG for coronary ischemia, it is widely recognized that arrhythmias are often excluded due to their potential to complicate signal interpretation. While these exclusions limit the broader applicability of the findings, they may enhance diagnostic accuracy by reducing confounding factors in signal analysis.

Diagnosing ischemia in patients with left bundle branch block (LBBB) presents a well-documented challenge, even with standard 12‑lead ECGs. The introduction of the Sgarbossa criteria in 1996, followed by subsequent revisions, highlights the difficulties in interpreting ECGs in this patient subset. Similarly, as MCG technology evolves, we anticipate the development of criteria akin to the Sgarbossa framework, specifically tailored for MCG signals in the presence of LBBB.

Looking ahead, the continued advancement of MCG technology is likely to drive the creation of more sophisticated interpretation methods. These innovations will enhance the assessment of ischemia in patients with arrhythmias and conduction abnormalities, broadening the clinical utility of MCG.

The definition of ischemic chest pain, based on positive functional testing, coronary angiography, or elevated troponin I, raises concerns. While perfusion defects and significant luminal stenosis (>70 %) are diagnostic, type II NSTEMI and other causes of elevated troponin could introduce bias. A standardized definition of ischemic chest pain would improve clarity and reliability.

The definition of ischemic chest pain, based on positive functional testing, coronary angiography, or elevated troponin I, raises valid concerns. While perfusion defects and significant luminal stenosis (>70 %) may be diagnostic, Type II NSTEMI and other causes of elevated troponin could introduce bias. A standardized definition of ischemic chest pain would improve clarity and reliability. It is however, important to recognize that all of these do indeed represent clinically utilized diagnostic approaches and all lack some specificity.

We appreciate this observation and agree that a Type II NSTEMI may indeed result in positive MCG findings. However, we do not consider this to be clinically inappropriate. This study aimed to evaluate patients presenting with chest pain who underwent clinical workup for ischemia. Patients were classified as having ischemic chest pain if their standard of care evaluation revealed abnormal results from functional testing or coronary angiography, defined as elevated troponin I, perfusion defects on stress testing, or ≥ 70 % luminal diameter stenosis and major epicardial vessels.

While troponin I was part of the clinical workup to assess coronary ischemia, it is a blood biomarker (currently superseded by high-sensitivity troponin) and not a functional test or coronary angiography.

Ischemia, by definition, occurs when the oxygen supply is insufficient to meet the metabolic demands of the myocardium, leading to impaired cellular function. Regardless of the underlying cause—whether due to a Type I or Type II NSTEMI—the physiological consequence is myocardial ischemia, which MCG is designed to detect. Identifying ischemia, irrespective of its etiology, aligns with the clinical utility of MCG as a tool for evaluating the myocardial oxygen supply-demand balance.

From a clinical perspective, recognizing ischemia in the context of Type II NSTEMI is equally critical, as it necessitates additional diagnostic evaluation and appropriate management to address the underlying cause and mitigate further myocardial injury. Therefore, MCG's ability to detect ischemia in such scenarios supports its relevance and value in diverse clinical contexts.

To the extent that the focus is specifically on identifying symptomatic and clinically significant coronary luminal stenosis, non-invasive functional testing alone may be insufficient due to known false positives and false negatives. Adjudicating equivocal angiographic findings of luminal stenosis would require intracoronary physiologic measures such as FFR or iFR. In this study, half of the patients underwent coronary angiography, and FFR or iFR was employed at the clinician's discretion.

The abstract mentions that all patients were “chest pain-free” at the time of scanning. However, this seems contradictory to the study's objective of evaluating MCG for ischemic chest pain. Clarification is needed on whether being chest pain-free was a predefined criterion and how this aligns with the study's goals. Additionally, it should be clarified whether having chest pain at the time of the study would have been grounds for exclusion from the study.

We agree that this aspect could be confusing and may lead to misinterpretation. It is our understanding that the authors were referring to patients who had initially presented with signs and symptoms of chest pain suspicious for coronary ischemia. However, at the time of their MCG scans, their chest pain symptoms were not actively present.

The intermittent nature of chest pain, even in cases of coronary ischemia, is not unusual—in fact, it is the norm. Clinicians frequently encounter this phenomenon in practice. Patients may experience episodic chest pain due to transient ischemic events, which resolve spontaneously or with medical intervention.

The primary objective of evaluating these patients is to identify and address reversible ischemia before it progresses to an irreversible myocardial infarction. This approach aligns with the study's goal of assessing MCG's utility as a diagnostic tool for detecting ischemia, even when chest pain is not present at the time of evaluation.

To clarify, it would be helpful for the authors to explicitly state whether being chest pain-free at the time of scanning was a predefined criterion and whether active chest pain at the time of the study would have led to exclusion.

The MCG system uses seven parameters, based on magnetic vectors during the Twave of the heart's electrical cycle, to detect ischemia. These parameters were defined using data from 197 patients who underwent tests like troponin, stress tests, and coronary angiography. If any of the parameters fall outside the normal range, the patient is classified as ischemic. [2] However, a major concern is that these criteria have not been validated in larger patient populations. To improve reliability, the next step should be to validate these criteria in a larger cohort using gold standard tests, such as coronary angiography, similar to how ST elevation in the right context on an ECG is validated.

There is a substantial body of literature describing abnormalities in the repolarization phase in patients with coronary ischemia, observed in both ECG and MCG studies. Both modalities measure the heart's electrophysiological activity and are capable of detecting abnormalities in depolarization and repolarization. However, the MCG signal is inherently more sensitive due reduced tissue attenuation by not requiring skin contact, making it unaffected by variations in body composition and conductivity that can influence ECG readings and MCG sensors detecting the heart's magnetic waves in both planar and tangential directions rather than an ECG's capture of voltage in just a two-dimensional plane [3,4].

That said, we agree that further validation studies are essential to establish the reliability and clinical utility of MCG. Specifically, studies comparing MCG against widely accepted non-invasive functional modalities, such as stress echocardiography or MPI-SPECT, and benchmarked against gold-standard tests like coronary angiography with intracoronary physiologic measures (e.g., FFR or iFR) or coronary PET, are necessary.

Such research would provide a robust framework for confirming the diagnostic accuracy of the MCG parameters and their application in diverse clinical settings. Until these larger, tightly controlled studies are performed, MCG remains a highly interesting and promising diagnostic tool that warrants further investigation and refinement.

Another issue is the equal weighting of all seven parameters. Treating them all equally may not reflect their true clinical importance, as some parameters might be more strongly associated with ischemia than others. This could be further studied in larger cohorts, which would also help refine the diagnostic performance.

We completely agree that treating all seven parameters with equal weighting may not accurately reflect their true clinical significance, as some parameters might be more strongly associated with ischemia than others. Further analysis in larger cohorts is essential to validate and refine the significance of these parameters.

Additionally, it is possible that there are other parameters beyond the seven evaluated and restricted to the repolarization phase in this study that could contribute to improving the diagnostic accuracy of MCG. More comprehensive research could help identify these additional parameters and better determine their role in detecting ischemia.

An approach utilizing AUC-ROC analysis or other advanced statistical methods could be instrumental in distinguishing parameters that enhance sensitivity from those that improve specificity. Such efforts would be valuable in optimizing the diagnostic performance of MCG, making it an even more effective tool for clinical use.

The study reports a sensitivity of 76.4 %, specificity of 74.3 %, positive predictive value (PPV) of 70 %, and negative predictive value (NPV) of 80 % for detecting ischemia. While the MCG performed better than the 12‑lead ECG (accuracy: 75.2 % vs. 61.6 %), there is still room for improvement. This performance gap may be due to the oversimplification of the parameters, which have not been thoroughly studied yet. This suggests that further research and validation are necessary to enhance the MCG system's accuracy.

We agree that further research and validation have the potential to significantly enhance the diagnostic accuracy and clinical utility of MCG. Refining and expanding the analysis beyond the seven parameters currently utilized could be a key step in this process. For example, incorporating additional features beyond the T-wave, such as elements of the ST segment, may provide valuable diagnostic insights and improve the system's performance.

Further research should also focus on tailoring the analysis to the clinical context. Depending on the specific use case, adjustments could be made to optimize sensitivity by prioritizing certain features or to enhance specificity by increasing the number of positive features required for classification. These modifications, guided by advancements in ROC curve analysis, could improve MCG's diagnostic performance and align it more closely with the needs of clinical practice.

Including presumed healthy controls increased the specificity and negative predictive value (NPV) to 89 % and 94 %, respectively. However, as mentioned earlier, without validating these controls through functional testing, this approach may overestimate diagnostic accuracy. The study found that MCG achieved 100 % sensitivity and NPV in low-to-intermediate risk patients but performed less effectively in high-risk patients. This raises the question of whether the criteria for defining ischemia via MCG need to be reexamined and refined.

We agree with both points raised. The inclusion of “healthy controls” based on criteria such as being asymptomatic, non-smoking, and without known hypertension, diabetes, or heart disease is a pragmatic and clinically acceptable approach. However, the lack of definitive confirmation does present limitations in terms of scientific rigor.

As previously noted, the authors addressed this concern by performing two separate calculations for the Negative Predictive Value (NPV). In the first calculation, they excluded healthy controls, resulting in an NPV of 89 %. In the second, they included healthy controls, increasing the NPV to 94 %. By presenting both figures, the authors provide a transparent and balanced view of their data, allowing readers to understand how the inclusion of healthy controls impacts the findings. This dual approach enhances the study's interpretability and demonstrates a commitment to transparency.

It is also worth reiterating that functional stress testing, which is widely accepted as part of clinical practice, is not without its limitations, as false positives and negatives are known to occur.

The higher NPV observed in low-to-intermediate-risk patients aligns with statistical expectations. In populations where the prevalence of the condition is lower, a higher NPV is naturally observed, whereas in higher-risk populations, this metric tends to decrease.

Finally, the decision to classify an MCG as ischemic if even one of the seven equally weighted parameters falls outside the normal range is notable. While this approach likely increases sensitivity, it may reduce specificity; however, it does provide excellent screening potential. Adjusting the threshold—for instance, requiring more than one parameter to be abnormal—could enhance specificity, depending on diagnostic and clinical priorities. Striking a balance between sensitivity and specificity and allowing clinicians the flexibility to prioritize one over the other based on the clinical scenario, offers a promising pathway to further refine the utility of MCG as a diagnostic tool.

The performance of MCG in patients with prior IHD (sensitivity 86 %, specificity 90 %, PPV 94 %, NPV 77 %) also warrants further exploration. It remains unclear how myocardial scarring, common in patients with prior MI, may have influenced the results. Given MCG's previous role in assessing myocardial viability, [3] distinguishing between scarring and acute ischemia could improve its clinical utility. Future studies should address this gap by delineating criteria to differentiate chronic scars from acute ischemic changes.

Myocardial scarring is common in patients with a prior history of MI. The potential influence of myocardial scarring on MCG results is an important consideration. We agree that given MCG's role in assessing myocardial viability, differentiating between scarring and acute ischemia is essential. Similar to how ECG differentiates between fixed and reversible defects based on electrophysiological changes and corresponding ECG patterns, MCG is expected to perform similarly, potentially with even higher sensitivity due to factors previously discussed. Nonetheless, a subgroup analysis with a larger sample size would be valuable. We agree that future studies should focus on delineating criteria that better distinguish chronic scars from acute ischemic changes.

It is also important to note that 86 % of the study's patients underwent stress SPECT or coronary angiography, which would ideally identify fixed defects associated with myocardial scarring. Comparing the results of these diagnostic modalities with MCG findings could offer insights into how well MCG aligns with standard-of-care interpretations. Nevertheless, additional research is warranted to address this gap and refine the role of MCG in this context.

The study presents an exciting development in MCG as a non-invasive tool for detecting ischemia. While the findings are promising, further large-scale studies are needed to refine the accuracy and standardize the diagnostic criteria. This is an excellent study, and we look forward to continued research to optimize and validate this innovative technology.

We agree with your conclusion that MCG is a highly promising non-invasive diagnostic tool, and it is exciting to witness the growing interest and potential applications for this technology, which has long been underexplored. Recent advancements in magnetometer sensor technology, coupled with significant improvements in computing power, have provided new opportunities for MCG to emerge as a valuable tool in clinical diagnostics. These technological strides have reignited interest in MCG, highlighting its ability to capture subtle electrophysiological changes in the heart, offering advantages that could help overcome some of the challenges associated with current traditional modalities. The efficiency and safety associated with MCG are particularly appealing in clinical practice.

As we continue to explore the full capabilities of MCG, it is essential to delve deeper into its clinical applications, ranging from ischemia detection to a variety of other potential uses. The versatility of MCG, combined with its non-invasive nature, positions it as a compelling option for numerous clinical scenarios, with the potential to significantly improve patient outcomes by enabling earlier and more accurate diagnoses.

However, to fully unlock its potential, larger-scale clinical studies and more focused investigations are necessary. These efforts will help refine its diagnostic accuracy, better define the clinical settings in which MCG excels, and establish standardized criteria that ensure its optimal use in practice. We look forward to the continued progress of this exciting field and the ongoing research that will undoubtedly contribute to MCG's development as an essential tool in cardiovascular diagnostics.

CRediT authorship contribution statement

Robert B. Takla: Writing – original draft.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.