MRI Evaluation of Neuromas in Oncological Amputees
Floris V. Raasveld, MD, Eva van Vliet, MD, Wen-Chih Liu, MD, Mark E Fleming, DO, FAAOS, Ian L Valerio, MD, MS, MBA, Kyle R. Eberlin, MD, Erik T. Newman, MD, Mohamed Jarraya, MD, Frank J. Simeone, MD and Jad Husseini, MD, Massachusetts General Hospital/Harvard Medical School, Boston, MA
Introduction: Tumors in the extremities may require amputation for oncologic cure, however, symptomatic neuroma following amputation is a primary cause of neuropathic pain. It remains unknown to what extend time affects neuroma growth and whether morphology influences pain intensity. Therefore, we aim to assess whether time influences radiographic neuroma size and shape, and their association with pain intensity in oncological amputees.
Methods: Oncological patients who underwent traditional extremity amputation without adjunctive nerve techniques such as Targeted Muscle Reinnervation (TMR) or Regenerative Peripheral Nerve Interface (RPNI) were included. MRIs were assessed post-amputation and prior to potential neuroma surgery. Pain of the residual limb, expressed on a Numeric Rating Scale (NRS, 0-10) within 3 months of the MRI assessment, and presence of neuropathic pain symptoms, were collected from chart review. Neuromas were classified as symptomatic neuromas or non-symptomatic neuromas based on description of pain with neuropathic symptoms. Two musculoskeletal radiologists independently assessed the MR images and described MRI features for each neuroma. Any discrepancy was solved by a third musculoskeletal radiologist serving as independent adjudicator. Neuroma size was expressed as the neuroma-to-nerve-ratio (NNR) (Fig. 1).
Results: Sixty patients were included in this study, of which 27 (45.0%) were female and 49 (81.7%) were lower extremity amputees. A total of 78 neuromas were identified on MRI, of which 56 (71.8%) were symptomatic with a median pain score of 3.5 (IQR 2.0-5.0). The median radiographic NNR of all neuromas was 2.0 (IQR 1.5-2.9). NNR was not significantly different between symptomatic and asymptomatic neuromas (p=0.665), and varying pain intensity (p=0.471) (Fig. 2). A larger NNR was associated with longer time-to-neuroma excision interval (p=0.011, Spearman’s ρ=0.285) (Fig. 3), and a smaller proximal nerve width (p<0.001, Spearman’s ρ=-0.596) (Fig. 4). Symptomatic neuromas were associated with upper extremity amputation (p=0.001), T2 heterogeneity (p<0.001) and presence of distal heterotopic ossification of the adjacent residual bone (p=0.029). T2 heterogeneity (p=0.003), perineural edema (p=0.031), enlarged fascicles (p=0.042) and presence of distal heterotopic ossification of the adjacent bone (p=0.024) were associated with more painful neuromas.
Conclusion: As assessed on MRI, smaller nerve caliber was associated with a larger NNR, and radiographic neuroma size increased with time. Size (NNR) did not influence neuroma pain intensity. MRI features associated with neuroma pain intensity were identified, and these findings may aid in physician awareness of neuromas in the oncological population and patient selection for surgical neuroma treatment.
Back to 2024 Abstracts