Diagnostic Utility of N-Terminal Pro-B-Type Natriuretic Peptide in Identifying Atrial Fibrillation Post-Cryptogenic Stroke: A Systematic Review and Meta-Analysis

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Dear Authors,

I read the manuscipt with great pleasure. The topic is extremely interesting and concern diagnostic utility of N -Terminal pro B -type Natriuretic Peptid in identifying atrial fibrillation post cryptognic stroke. More than 2/3 of cryptogenic strokes are embolic, mainly cardiogenic. That is why cardiac imaging and the diagnosis of cardiac arrhythmias are so important, especially atrial fibrillation. Indeed, the concentration of NTproBNP increases rapidly after the onset of atrial fibrillation. 

The methodology and results are described in a clear and transparent manner.

The combination of NT-proBNP in blood serum, ECG telemetry, clinical factors, imaging tests, including transthoracic echocardiography and transesophageal echocardiography may allow to detect the cause of stroke. The authors emphasized this fact in the discussion and conclusions.

Figure 2 and 3 are barely visible even the text is enlarged.

Best regards 

Author Response

Dear Reviewer,

We would like to thank you for comprehensive review of our work and encouraging feedback on our manuscript. We have made best of efforts to incorporate these comments. In addition, a revised version of the manuscript with track changes in red is attached.

In response to your comments, we would like to provide the following rebuttal and clarifications:

1. Importance of Cardiac Imaging and Diagnosis of Cardiac Arrhythmias: We are glad that you highlighted the importance of cardiac imaging and arrhythmia diagnosis in cases of cryptogenic stroke. Our study emphasizes the rapid increase in NT-proBNP concentration following the onset of atrial fibrillation, which reinforces the utility of this biomarker in early detection and diagnosis.　We indeed have discussed this in detail both in Introduction as well as in Discussion. The discussion has been expanded as per the suggestions.
2. We appreciate your positive remarks on the clarity and transparency of our methodology and results. 
3. We concur with your observation that combining NT-proBNP measurements with ECG telemetry, clinical factors, and imaging tests like transthoracic and transesophageal echocardiography is critical in identifying the cause of stroke. Our discussion and conclusions indeed emphasize this integrated approach, underscoring its potential to enhance diagnostic accuracy. The conclusion has been updated as below: 

Page 18: In conclusion, our study emphasizes the rapid increase in NT-proBNP concentration following the onset of atrial fibrillation, reinforcing the utility of this biomarker in early detection and diagnosis. Our meta-analysis found that NT-proBNP demonstrates good to very-good diagnostic accuracy in detecting AF following cryptogenic stroke, with excellent accuracy observed in patients with stroke of known etiology. This meta-analysis updates and consolidates the data to assess the diagnostic accuracy of NT-proBNP in detecting AF in the context of stroke. An integrated approach combining NT-proBNP measurements with ECG telemetry, clinical factors, and imaging tests like transthoracic and transesophageal echocardiography is critical in identifying the cause of stroke and enhances diagnostic accuracy. Although consensus on a standardized risk tool is still pending, from a clinical perspective, NT-proBNP holds promise in identifying cryptogenic stroke patients who would benefit from extended cardiac monitoring. Further primary research is warranted to determine the optimal cut-off for NT-proBNP and to identify the most effective form of cardiac monitoring for these patients.

4. Figures 2 and 3: We apologize for the visibility of Figures 2 and 3. We attempted to revise these figures to improve their clarity and readability; however, STATA offers limited options for editing the figures. We are providing the high-resolution version of the figures generated. We also recognize this, and the output of the figure has also been tabulated in Table 4, which we hope could suffice.

Thank you once again for your constructive feedback. We hope that the revised manuscript will meet your expectations.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The manuscript is well written. The issue addressed in important. The results are as expected.

Minor issues to be clarified.

How was cryptogenic stroke diagnosed in each study? What was the preceding word up (TOE, etc)? 

When was AF diagnosed in each study post enrollment? In general monitoring was not intense in most studies?

What might be the cut-off of NTproBNP for predicting AF post stroke? (250, 500)

Figures 2 and 3 are not clear. You should improve analysis and quality.

Author Response

Dear Reviewer,

Thank you for your constructive feedback and for recognizing the significance of our work. We appreciate your detailed comments and have addressed each point as follows. In addition, a revised version of the manuscript with track changes in red is attached.

1. “How was cryptogenic stroke diagnosed in each study? What was the preceding word up (TOE, etc)?”

Reply# In our revised manuscript, we have elaborated on the diagnostic work-up of cryptogeic stroke for each study, where available, in the Table 1 (please see new column added)

2. When was AF diagnosed in each study post enrollment? In general monitoring was not intense in most studies?

Reply# We have now included information on the detection timeline for the diagnosis of atrial fibrillation (AF) post-enrollment in each study (please see Table 1 (column “AF detection method”), where the deta was available. We discuss the variability in monitoring intensity and its potential impact on AF detection rates. We have also added a paragraph to the discussion on the inconsistencies in AF diagnosis used across the studies, as below;

Discussion (Page 16): A limitation of our meta-analysis is that there were inconsistencies in the timing of AF diagnosis post-enrolment in each study, and the duration of cardiac monitoring varied. The most common approach to AF detection was an ECG on admission and 24-hour Holter monitoring; however, prolonged ECG monitoring outside the hospital was not always conducted. As a result, AF may have been underdiagnosed in some cases. Specific information on the AF detection strategies employed in each study is described in Table 1. In the era of reperfusion therapy, AF is an important clinical consideration as it mediates outcomes in acute ischemic stroke patients treated with intravenous thrombolysis [43]. This adds weight to the importance of NT-proBNP assessment, indicating that it may even be crucial in acutely prognosticating patients being treated with reperfusion therapy.

 

3. What might be the cut-off of NTproBNP for predicting AF post stroke? (250, 500)

Reply¢ We thank the reviewer for this valuable suggestion. We indeed agree that is this is an important consideration. A subsection has now been added to the Discussion/Limitations (Page 17), as below:

The variations in NT-proBNP cut-offs emerged as a major limitation of this study. There were considerable variations in NT-proBNP cut-offs in our study, ranging from 250 pg/mL to 505 pg/mL (Table 3). Theoretically, a lower cut-off (such as 250 pg/mL) would yield higher sensitivity for AF diagnosis but lower specificity, and vice versa for a higher cut-off (such as 505 pg/mL). Unfortunately, there were insufficient studies at these data points for us to conduct subgroup analyses, and determining the ideal cut-off remains an area for future primary research. It is also possible that there is no single “ideal” cut-off, as clinicians may benefit from considering patient demographics and resource availability to decide if a higher specificity or higher sensitivity approach is more suitable for their circumstances. Nevertheless, our data suggests that any NT-proBNP value within the range of 250 to 505 pg/mL provides reasonable sensitivity and specificity for AF detection post-stroke.

4. Figures 2 and 3 are not clear. You should improve analysis and quality.

Reply# Figures 2 and 3: We apologize for the visibility of Figures 2 and 3. We attempted to revise these figures to improve their clarity and readability; however, STATA offers limited options for editing the figures. We are providing the high-resolution version of the figures generated. We also recognize this, and the output of the figure has also been tabulated in Table 4, which we hope could suffice.

We hope these revisions address your concerns and enhance the manuscript's clarity and impact. Thank you again for your valuable feedback.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The authors responses are appropriate 

The manuscript can be approved