Postoperative Imaging Characteristics of Targeted Muscle Reinnervation and Regenerative Peripheral Nerve Interfaces

Abstract

Background:

Targeted muscle reinnervation (TMR) and regenerative peripheral nerve interface (RPNI) procedures can mitigate the incidence of neuroma; however, neuromas may still form. Interval soft tissue changes at the site of these nerve constructs may pose a diagnostic challenge. This review examined existing literature on ultrasound (US) and magnetic resonance imaging (MRI) characteristics of TMR and RPNI to aid in their identification.

Methods:

A literature review was conducted using the PubMed, Embase, Cochrane, and Web of Science databases following the preferred reporting items for systematic reviews and meta-analyses 2020 protocol to examine studies describing imaging characteristics of TMR and RPNI in humans. Studies were screened by 2 authors. Studies that described imaging characteristics of TMR and RPNI were included.

Results:

The review yielded 623 articles. Ten articles were identified for full-text review, and 4 met the inclusion criteria. One article described the US appearance of TMR. Two articles described the US characteristics of RPNI. One article described the appearance of RPNI on MRI. A total of 44 patients, 70 RPNIs, and 6 TMRs were included.

Conclusions:

The characterization of TMR and RPNI on US and MRI is limited, particularly regarding upper extremity TMR and RPNI. TMR and RPNI demonstrate consistent core features on US and MRI. When monitoring patients for pain postamputation, plastic surgeons should be aware of the expected changes to the nerve and surrounding soft tissues after TMR and RPNI to aid in clinical judgment of new soft tissue lesions identified on imaging after amputation.

Takeaways

Question: What are the ultrasound (US) and magnetic resonance imaging (MRI) characteristics of targeted muscle reinnervation (TMR) and RPNI based on the existing literature?

Findings: On US, successful TMRs demonstrate hypoechoic nerve fascicles merging into the target muscle. RPNIs display degenerative and regenerative changes at the nerve–muscle junction, within the muscle graft, and along the nerve. On MRI, RPNIs are isointense to muscle on T1-weighted images and hyperintense on T2-weighted images, with variable contrast enhancement.

Meaning: TMR and RPNI demonstrate consistent core features on US and MRI, which plastic surgeons should be aware of when monitoring patients for pain postamputation.

INTRODUCTION

A potential complication of amputation is painful neuroma formation after nerve transection.^1–3 Chronic pain is a common complication of amputation and has a significant negative impact on a patient’s quality of life. Up to 59% of amputation patients experience residual limb pain, with 15% of cases attributed to symptomatic neuromas.^1–7 Following injury, nerves undergo Wallerian degeneration, marked by Schwann cell death and distal axon degradation to the first node of Ranvier. Macrophages then clear debris, forming a decellularized endoneurial conduit that promotes Schwann cell proliferation and axonal regrowth toward a distal nerve target. Without a target, such as in an amputation, or if this process is obstructed by cellular debris or inflammatory fibrosis, axons grow in a disorganized manner, leading to spindle neuromas along the axon or terminal neuromas at the transection site.^8,9

Targeted muscle reinnervation (TMR) and regenerative peripheral nerve interface (RPNI) were originally developed to improve prosthetic control but are now increasingly used prophylactically after amputation to reduce neuroma incidence and as treatments for symptomatic neuromas.^3,10–18 TMR involves coapting a transected peripheral sensory or motor–sensory nerve to a functional motor nerve unit. RPNI involves wrapping a denervated muscle graft around a transected nerve to prevent nerve exposure at the stump while providing a target for neuronal regeneration into the muscle fibers.^{12,15,19–21} Both procedures reduce neuroma incidence by providing a target for transected nerves to regenerate into, thereby giving the nerve a function as it heals and reducing the risk of disorganized growth and neuroma formation.^3,7,12

Although magnetic resonance imaging (MRI) and biopsy are the gold standard for diagnosing neuromas in general, the imaging characteristics of TMR and RPNI constructs, as well as the optimal imaging modality to assess them, are not well characterized.^9,22,23 This poses a particular challenge in the setting of patients who undergo amputation due to malignancy and require routine evaluation for recurrence of soft tissue malignancies at the residual stump site.⁹ In patients with extremity sarcoma, local recurrence rates have been reported in 7%–36% of oncological amputation patients.^1,24–31 Following TMR or RPNI, amputation patients are routinely followed up to assess for the incidence of residual limb pain and undergo surveillance imaging to assess for tumor recurrence. Interval soft tissue changes at the site of these nerve constructs may pose an additional confounding variable and create the potential for misinterpretation between recurrence versus surgical change after TMR and RPNI or neuroma formation.⁹

This scoping review aimed to examine the existing literature on ultrasound (US) and MRI characteristics of TMR and RPNI to improve the identification of these constructs in postoperative imaging. Additionally, it sought to provide insights into differentiating benign postsurgical changes from findings that may require further intervention, such as neuroma formation or malignancy recurrence.

METHODS

This study was approved by our institutional review board (protocol 2023P002405). A scoping review was conducted due to the novelty of this area of research and the limited number of articles that met the inclusion criteria. During September 2024, published articles were comprehensively reviewed using the PubMed, Embase, Cochrane, and Web of Science databases, following the preferred reporting items for systematic reviews and meta-analyses (PRISMA) 2020 reporting guidelines and the JBI Manual for Evidence conducting guidelines to identify and examine studies that described MRI and US characteristics of TMR and RPNI.^32,33 This review was registered with the Open Science Framework registry (DOI: 10.17605/OSF.IO/F7ZW6). No deviations were made from the original protocol.

Search Strategy

A search strategy was completed using the following controlled vocabulary and keywords: (targeted muscle reinnervation[tiab] OR targeted muscle re-innervation[tiab] OR tmr[tiab] OR regenerative peripheral nerve interface*[tiab] OR rpni[tiab] OR rpnis [tiab]) AND (“Ultrasonography”[Mesh:NoExp] OR “peripheral nerves/diagnostic imaging”[mesh] OR ultraso*[tiab] OR sonograph*[tiab] OR “magnetic resonance imaging”[Mesh] OR magnetic resonance[tiab] OR mr image*[tiab] OR nmr[tiab] OR mri[tiab] OR mris[tiab] OR neurograph*[tiab]). All articles identified from the search strategy were populated into Covidence for analysis.³⁴

Eligibility Criteria

We conducted a scoping review to qualitatively summarize the imaging appearance of TMR and RPNIs, given the limited and heterogeneous literature available. We included primary studies written in English and published before September 2024 that described TMR or RPNI using US or MRI and contained representative images. Including only studies with images allowed us to directly verify and illustrate reported imaging findings, enhancing interpretability and clinical relevance. Eligible study designs included randomized controlled trials, cohort studies, case-control studies, case series, and case reports to capture all available evidence. In practice, only case series met the criteria, reflecting the early stage of research in this field. We excluded nonhuman studies, studies of neuromas in patients without TMR or RPNI, and studies lacking images of sufficient quality to confirm the described characteristics.

Study Selection

A 2-stage screening process was conducted for study selection. First, 2 reviewers (A.-A.M.) and (R.K.) independently screened abstracts that resulted from this search strategy using the predetermined inclusion and exclusion criteria. The reviewers were blinded to screening votes. Once abstracts were screened, a full-text review was conducted by the reviewers to confirm eligibility for inclusion. A third reviewer (L.A.H.) was used as a tie-breaking vote to resolve discrepancies between the 2 reviewers.

Data Extraction and Synthesis

Once studies meeting inclusion criteria were identified, qualitative meta-aggregation was performed to synthesize data on patient number, median age, amputation site, number and type of TMRs or RPNIs, nerves involved, imaging modality, postoperative imaging day, and imaging appearance on MRI or US.³² Data extraction was independently conducted by 2 reviewers (A.-A.M. and R.K.) to minimize bias and ensure accuracy. The same reviewers performed critical appraisal following the 2024 JBI Manual for Evidence Synthesis (Table 1).³⁹ A standardized data collection form was developed and pilot-tested for consistency. Clear reporting of patient demographics, inclusion of pertinent clinical history, clear reporting of imaging findings, description of patient follow-up, and conclusions supported by data were assessed as present or absent for each article. Articles that had all these parameters were considered high quality for the intents of this scoping review. Discrepancies were resolved by consensus. An executive summary of the included articles is presented in Table 2. The study population encompassed patients of all ages and ethnicities undergoing TMR or RPNI after amputation for any indication. Primary outcomes included imaging descriptions of nerve fascicle architecture, TMR/RPNI morphology, echogenicity relative to surrounding soft tissue and intact nerves, and MRI signal characteristics on T1- and T2-weighted sequences with or without contrast. Secondary outcomes examined correlations between imaging findings, successful TMR/RPNI, and neuroma recurrence. All results relevant to these outcomes were included for analysis.

Table 1.

JBI Critical Appraisal of Included Studies

Author, Year	Study Type	Patient Demographics Reported (Y/N)	Clinical History Provided (Y/N)	Diagnostic Modality Described (Y/N)	Imaging Findings Clearly Reported (Y/N)	Follow-up Described (Y/N)	Conclusions Supported by Data (Y/N)	Quality
Finkelstein et al 202335	Case series	Y	Y	Y	Y	Y	Y	High
Lin et al 202336	Retrospective analysis	Y	Y	Y	Y	Y	Y	High
Morag et al 202237	Retrospective analysis	Y	Y	Y	Y	Y	Y	High
Morag et al 202438	Retrospective analysis	Y	Y	Y	Y	Y	Y	High

Table 2.

Executive Summary of Included Articles

Article Title	Author, Year	No. Patients	Median Age, y	Amputation Site	No. TMR or RPNIs	Nerves	Imaging Modality	Median Postoperative Day of Imaging	Imaging Included
Experience with ultrasound neurography for postoperative evaluation of targeted muscle reinnervation	Finkelstein et al 202335	3	53	Upper extremity	6 TMR	2 median, 2 radial, 2 ulnar	US	13.5 mo	Yes
Regenerative peripheral nerve interface reduces the incidence of neuroma in the lower limbs after amputation: a retrospective study based on ultrasound	Lin et al 202336	7	25.9	Lower extremity	7 RPNIs	3 sciatic, 2 tibial, 2 CPN	US	3 mo	Yes
Ultrasound appearance of regenerative peripheral nerve interface with clinical correlation	Morag et al 202237	21	42	Lower extremity	45 RPNIs	41 CPN, 4 tibial	US	11 mo	Yes
Magnetic resonance imaging appearance of regenerative peripheral nerve interface	Morag et al 202438	14	51	Upper and lower extremity	18 RPNIs	8 sciatic, 5 tibial, 4 CPN, 1 brachial plexus	MRI	16.5 mo	Yes

CPN, common peronal nerve.

RESULTS

The initial search strategy yielded 623 articles. Ten articles were identified for full-text review, and subsequently, 4 articles were determined to meet the inclusion criteria. Collectively, these studies comprised 44 patients who underwent a total of 70 RPNI and 6 TMR procedures. The reported median age of patients ranged from 25.9 to 53 years, with most cases involving lower extremity amputations (41 of 44 patients, 93%). The included studies were published between 2022 and 2024, and all provided representative images of TMR or RPNI.

Regarding imaging modalities, 1 article described the US appearance of TMR.³⁵ Two articles described the US characteristics of RPNI.^36,37 One article described the appearance of RPNI on MRI.³⁸ There was only 1 published case of the MRI appearance of an upper extremity TMR.

On US, TMR was reported as hypoechoic nerve fascicles extending into the muscle. RPNI demonstrated variable appearance across 3 regions: the nerve–muscle junction, the muscle graft, and the nerve within the RPNI. On MRI, RPNIs were isointense to muscle on T1-weighted images and hyperintense on T2-weighted images. Postcontrast enhancement was reported in most cases. The median postoperative interval to imaging ranged from 3 to 16.5 months. Reported amputation sites and targeted nerves varied, with the most common involving the common peroneal nerve (n = 47), sciatic nerve (n = 11), and tibial nerve (n = 11). A flow diagram of included studies is shown in Figure 1, and a thematic analysis is presented in Table 3.

Fig. 1.

PRISMA flow diagram for included studies.

Table 3.

Thematic Analysis of Included Articles

Theme	Summary of Findings	Supporting Studies
1. MRI image characteristics	RPNIs were typically isointense to muscle on T1-weighted images and hyperintense on T2-weighted images, without changes in intensity throughout follow-up. Postcontrast enhancement was commonly described as peripheral, fascicular, linear, and distal stump focal. Enhancement was not associated with neuroma symptoms or revision	Morag et al 202438; Lin et al 202336
2. Fascicular architecture and nerve continuity on US	In successful TMR, hypoechoic nerve fascicles were continuous with the proximal nerve and integrated into muscle without signs of neuroma or muscle atrophy. In unsuccessful TMR, sonographic signs included discontinuity of fascicles, neuroma formation, and denervation atrophy	Finkelstein et al 202335
3. Morphological zones of RPNI on US	RPNIs showed 3 distinct regions on US: the nerve–muscle junction, muscle graft, and nerve within the muscle–nerve unit. Findings often did not correlate with pain, and echogenic changes in the muscle graft were not predictive of symptoms	Morag et al 202237; Lin et al 202336
4. Neuroma differentiation on US and MRI	Neuromas following traditional amputation often preserve fascicular structure. RPNI and TMR constructs can resemble these, but fascicle continuity and integration are key differentiators. MRI contrast enhancement is not specific for pathological changes	Morag et al 202438; Lin et al 202336
5. Image stabilization	Imaging was conducted 3–16.5 mo postoperatively. Stable imaging occurred after 3–4 mo. Unlike typical nerve regeneration, T2 hyperintensity persisted in RPNIs	Morag et al 202237; Morag et al 202438
6. Modality selection: US versus MRI	US was favored over MRI for evaluating TMR and RPNI due to superior resolution, ability to view fascicular architecture, and capacity to visualize construct viability and integration. MRI, although the standard for neuroma detection, was less sensitive in differentiating intact versus symptomatic RPNIs	Morag et al 202237; Lin et al 202336; Finkelstein et al 202335

DISCUSSION

A contextual understanding of normal and pathological nerve imaging is necessary to interpret the unique features of TMR and RPNI. US provides a detailed assessment of the nerve structure and surrounding tissues. On US, individual nerve fascicles can be appreciated on the long axis view as hypoechoic structures in continuity with the proximal nerve and separated by hyperechoic bands reflecting the epineurium.^40,41 In the short-axis view, nerve fascicles surrounded by epineurium may have a honeycomb appearance. In acute nerve injury, US may show focal fascicular hypoechoic swelling or loss of the normal honeycomb pattern due to interfascicular edema.⁴² In chronic cases, epineurial and intraneural fibrosis can cause hyperechogenic changes, disrupting the honeycomb appearance or altering nerve caliber.⁴³

Following standard amputation without TMR or RPNI, neuromas typically appear on US as hypoechoic enlargements continuous with the proximal nerve, but morphology is variable. However, their appearance can vary significantly and has been described using terms such as “bi-lobed nodules,” “pseudocystic ovoid masses,” “fusiform masses with a radiating network of smaller nerves,” and “cylindrical enlargements of normal nerve morphology.”^23,44 In addition, the honeycomb fascicular morphology of the nerve may be preserved in a neuroma, as described by Sandomenico et al⁹ as a “Hypoechoic fusiform mass with echogenic strands inside a bulbous end continuous with normal nerve.” Although TMR and RPNI can show varied imaging appearances, they share similar US features due to the conserved anatomy of peripheral nerves. Therefore, sonographic assessments can detect changes associated with denervation and reinnervation of nerves postoperatively.³⁸

Although TMR and RPNI have been shown to reduce neuroma formation and phantom limb pain, neuromas may still develop in cases where these procedures are unsuccessful.^11,45,46 Finkelstein et al³⁵ highlighted sonographic differences between successful and unsuccessful TMR. Finkelstein et al³⁵ demonstrated that in successful TMR, US reveals hypoechoic nerve fascicles continuous with the target motor nerve and eventual integration into the target muscle; in unsuccessful TMR, US reveals discontinuity of nerve fascicles, neuroma formation, and signs of muscle denervation atrophy, such as decreased thickness, loss of striation, adipose infiltration, and increased echogenicity. These findings can appear as early as 2 weeks to 1 month post-TMR, with imaging stability typically occurring after 3–4 months (Table 3).

RPNIs exhibit variable appearances due to degenerative and regenerative processes that occur in the nerve–muscle junction, muscle graft, and nerve within the RPNI.³⁷ Morag et al³⁷ defined morphological changes into 7 groups. Key findings included preserved RPNI appearance; hypoechoic changes at the nerve–muscle junction; various swelling patterns with either intact or disrupted fascicular structure; and hypoechoic, mass-like appearances suggestive of neuroma. The appearance of muscle grafts varied due to surgical technique, with most demonstrating hypoechoic or isoechoic compared with surrounding fat. These findings are summarized in Table 3.

Postoperative observations showed muscle graft atrophy and potential echogenic changes indicating fatty infiltration or fibrosis. Thin echogenic structures between the nerve and muscle grafts were noted, likely representing connective tissue proliferation or disorganized nerve growth, though they did not correlate with pain symptoms.³⁷ In 8 patients, follow-up imaging conducted 4 months after initial imaging (11 mo postsurgery) showed no changes in RPNIs, indicating stabilization of these constructs. The study concluded that focal changes between the distal nerve stump and muscle, along with diffuse nerve swelling and loss of fascicular structure, may be linked to pain and suggest a need for revision surgery. Notably, even RPNIs with preserved architecture and without hypoechoic changes did not always align with the absence of pain, indicating that chronic pain can occur without distinct US findings. Additionally, echogenic changes in the muscle did not relate to persistent symptoms after RPNI.³⁷

Lin et al³⁶ characterized the US appearance of RPNI by comparing US at 3 months postoperatively in 7 patients who underwent RPNI after lower extremity amputation to 7 patients who underwent traditional amputation without RPNI. They found that RPNI significantly reduced the size of terminal neuromas compared with the control group and was correlated with significantly lower pain scores.³⁶

Although MRI is widely regarded as the gold standard for diagnosing neuromas in general, Lin et al,³⁶ Finkelstein et al,³⁵ and Morag et al³⁷ recommend US over MRI for the evaluation of TMR or RPNI due to its higher spatial resolution for superficial nerves, greater sensitivity to assess fascicular architecture of small nerves, ability to view the trajectory of nerves into the muscle, and ability to distinguish the components of the construct and judge its survival.^9,22 Thus, although MRI remains the standard for neuromas broadly, US may provide unique advantages in the evaluation of TMR and RPNI constructs, particularly in assessing their dynamic features and structural integrity, and can serve as a powerful adjunct modality in this context.

Nonetheless, in a subsequent study, Morag et al³⁸ described the MRI appearance of RPNIs in 14 patients following 13 lower extremity amputations and 1 upper extremity amputation, with MRIs obtained at 16.5 months after surgery. On MRI, RPNIs appeared isointense to muscle on T1-weighted sequences and hyperintense on T2-weighted sequences without interval change over time.³⁸ These findings lie in accordance with previous studies on the MRI characteristics of neuromas in patients who underwent standard amputation.^9,22,47 In addition, heterogeneity on T2-weighted images has been shown to be a possible predictive indicator for a symptomatic neuroma.²² The hyperintense T2 signal of RPNIs observed by Morag et al³⁸ is expected after peripheral nerve damage, as the degeneration of axons and myelin results in increased endoneurial edema. However, Morag et al³⁸ demonstrated that RPNIs do not demonstrate the expected resolution of T2 signal that occurs over time as the nerve regenerates and extracellular edema resolves, suggesting that the persistent increase in T2 signal may be due to a different mechanism in RPNIs.³⁸

Postgadolinium contrast enhancement is recognized as a useful indicator for identifying neuromas on MRI, as 78%–88% of neuromas after standard amputation demonstrate contrast enhancement.^22,48 Interestingly, Morag et al³⁸ also observed contrast enhancement in all but 1 RPNI and categorized the enhancement into 4 patterns: peripheral enhancement, punctate fascicular enhancement, plate-like/linear interfascicular enhancement, and focal enhancement at the distal nerve stump.³⁸ Most RPNIs exhibited a combination of 2–3 enhancement patterns, likely reflecting differential enhancement of the RPNI components. However, they did not find an association between contrast enhancement and the need for RPNI revision due to a suspected neuroma. This suggests that contrast enhancement may not be specific enough to distinguish RPNI from a symptomatic neuroma.

LIMITATIONS

To our knowledge, this is the first review of the imaging characteristics of TMR and RPNI, and given the novelty of this research topic, this study is not without limitations. A scoping review was conducted due to the novelty of this area of research and the lack of articles that met the inclusion criteria. In accordance with scoping review methodology and reporting standards (PRISMA), a formal risk of bias assessment was not performed, as the primary objective was to characterize the range and nature of evidence on the imaging appearance of TMR and RPNI.³³ We took several steps to minimize bias in our review process. These included performing comprehensive database searches, using predefined inclusion and exclusion criteria, and involving multiple independent reviewers in article screening and full-text review to enhance the reliability and objectivity of study selection.

To mitigate scope creep, we included only articles that specifically described the radiological appearance of TMR and RPNI and not neuromas in general. Additionally, because we excluded studies that did not include representative images of sufficient quality, our findings may be subject to selection bias, as potentially relevant studies without adequate imaging for verification were not included. Only 4 studies met the inclusion criteria for this review, reflecting the nascent nature of this field of research. This small sample size limits the strength and generalizability of our conclusions, as the findings are based on a very limited and heterogeneous evidence base. As such, our results should be interpreted as preliminary and largely descriptive, highlighting the need for further research with larger sample sizes and a broader range of imaging modalities and clinical settings. Moreover, as these are relatively new surgical procedures with an ever-evolving body of literature, some articles published after our scoping review was completed may not be included in this analysis. Future directions include further characterization of the imaging appearance of TMR and RPNI in larger patient series, including but not limited to characterization of TMR on US and MRI, characterization of upper extremity TMR and RPNI, and changes over time.

CONCLUSIONS

Characterization of the appearance of TMR and RPNI on US and MRI is limited, particularly with regard to TMR and the radiological evaluation of upper extremity TMR and RPNI. Therefore, there is a significant need for larger studies on the radiological appearance of these constructs. Although there is significant variability, TMR and RPNI demonstrate consistent core features on US and MRI. These imaging modalities can be used to appreciate changes in the anatomical components of these surgical constructs, which can be helpful in the identification of neuroma after TMR or RPNI. Radiologists and plastic surgeons should be aware of the expected changes that may occur to the nerve and surrounding soft tissues after TMR and RPNI to aid in their clinical judgment of new soft tissue lesions after extremity amputation.

DISCLOSURE

The authors have no financial interest to declare in relation to the content of this article.