Magnetic Resonance Neurography Findings in Clinically Suspected Posterior Interosseous Neuropathy: Response

Authors’ Response:

We sincerely thank the correspondent for the thoughtful comments and for recognizing the clinical significance of our study. We fully agree that the lack of standardized imaging thresholds represents an important limitation in magnetic resonance neurography (MRN), especially in the subjective assessment of nerve hyperintensity. To mitigate potential variability, our study incorporated independent evaluations by 2 musculoskeletal radiologists. Prior to grading, both readers participated in a consensus session to review the grading criteria and scales using separate image data sets that were not part of the study analysis. This approach ensured consistent interpretation and led to substantial to almost perfect interrater agreement across all imaging features, as detailed in Table 1 of our paper.

While universally accepted quantitative criteria could improve reproducibility, achieving such standardization remains challenging given the variety of MRN protocols, sequence parameters, and scanner hardware across institutions. We agree that multicenter collaboration will be essential to establish consistent diagnostic benchmarks. Recent consensus initiatives, such as the Neuropathy Score Reporting and Data System, have already proposed structured and standardized reporting for peripheral neuropathy, with multicenter validation supporting its reproducibility and clinical applicability.^1,2 Nonetheless, MRN should be seen as a powerful adjunct—rather than a standalone diagnostic test—that complements clinical exams and electrodiagnostic tests in routine practice. Its main benefit is providing anatomical correlation and pinpointing nerve pathology, thereby enhancing diagnostic confidence and guiding appropriate treatment planning.

Regarding clinical stratification factors, we concur that imaging appearance may vary among demographic subgroups. To minimize potential bias, we performed age matching within 3 years between the posterior interosseous nerve and control cohorts, as detailed in the Methods section. We also acknowledge that age-related fatty infiltration and muscle atrophy can confound the interpretation of denervation changes. To address this, our grading system specifically emphasized disproportionate fatty infiltration of the extensor musculature, focusing on asymmetry or localized patterns inconsistent with generalized age-related changes.

Regarding the concern over not considering contributions of alternative magnetic resonance imaging sequences for nerve evaluation, we did evaluate sequences with different contrast weightings. Proton density–weighted images were used to evaluate perineural fat planes and the arcade of Frohse, while T2-weighted Dixon sequences provided optimal contrast for detecting nerve signal hyperintensity and muscle denervation. The suggestion to analyze the “relative contribution” of each sequence appears to conflate our study with the cited work, which focused on using artificial intelligence or less experienced readers to identify necessary sequences for protocol optimization. Such studies focus on workflow efficiency and imaging protocol audits, assessing the number of acquired or omitted sequences, rather than diagnostic interpretation. Assessing any single sequence independently would not reflect standard MRN methodology or provide meaningful diagnostic information.

Finally, we fully concur with the correspondent's emphasis on interdisciplinary collaboration. At both the institution where the study data were collected and the corresponding author's (Y.L.) institution, multidisciplinary peripheral nerve conferences are held regularly, bringing together specialists in radiology, peripheral nerve surgery, neurology, and physiatry to foster collaborative decision making and coordinated patient care. We strongly encourage other centers with similar expertise and interest in peripheral neuropathy to adopt this collaborative framework, which we believe is essential for advancing diagnostic accuracy and improving patient outcomes.

Yenpo Lin, MDNew York, USATaoyuan, Taiwan Anisha Javvaji, BS New York, USA Vijay B. Vad, MD New York, USA Darryl B. Sneag, MDNew York, USA