Consecutive assessment of recovery after peripheral nerve injury of the sciatic nerve within the same rat using PET/MRI

Abstract

Background

Positron emission tomography (PET) has been reported as effective in diagnosing peripheral nerve injury (PNI). However, there is a lack of studies evaluating different degrees of PNI using PET within the same individual to reduce errors due to interindividual differences.

Purpose

To evaluate the recovery process in the same rat after sciatic nerve injury using PET/magnetic resonance imaging (MRI).

Material and Methods

Crushing nerve injuries were induced in the left sciatic nerves of six male rats, preserving the right ones. The degree of nerve damage was measured at one, two, three, four, and five weeks postoperatively using three assessment methods: paw withdrawal threshold test (RevWT); PET (SUVR); and MRI (MRSIR). All the representing values of each method are presented as ratio values of the right and left sides in each rat.

Results

Significant gradual recovery of all rats was observed over time in all the methods. No significant differences in RevWT and MRSIR were observed between before and more than four weeks after injury, whereas a significant difference in SUVR was still observed between before and five weeks after injury (P = 0.0007). The parameters of all methods decreased significantly over time (P = 0.000, all), and the explanatory power was significant in RevWT, SUVR, and MRSIR.

Conclusion

PET and MRI could be valuable non-invasive techniques for diagnosing neuropathic pain resulting from PNI. PET/MRI would be expected to be a more accurate and informative diagnostic tool for PNI than MRI alone.

Keywords

Peripheral nerve injuries peripheral neuropathies diagnosis positron emission tomography magnetic resonance imaging

Introduction

Neuropathic pain, characterized by sensory abnormalities such as dysesthesia, hyperalgesia, and allodynia, can be caused by peripheral nerve injury (PNI) from various causes, including trauma, inflammation, and chemical damage (1,2). In the oral and maxillofacial regions, the trigeminal nerve territories are often affected, and inferior alveolar and lingual nerve injuries most frequently occur (3). In the Republic of Korea, dental implants have caused the most medical disputes over PNIs during the last five years according to the statistics of Korea Medical Dispute Mediation and Arbitration Agency, and this trend might continue to grow as the number of patients requiring dental implant treatment increases.

Prompt and accurate diagnosis of nerve injuries can not only solve the medical disputes, but also lead to effective treatment and good prognosis. However, to date, the assessment modalities for PNIs in clinics, such as clinical neurosensory tests, current perception threshold, and digital infrared thermal imaging, may be subjective and have some limitations in evaluating the nerve conditions objectively (2,4,5). Consequently, current and prospective non-invasive imaging techniques, such as computed tomography, magnetic resonance imaging (MRI), ultrasonography (US), positron emission tomography (PET), and other molecular imaging techniques, have been introduced in several publications (4,6,7). Among them, PET, which can visualize metabolism and detect a disease early with high sensitivity, is suggested to have the most potential for diagnosing peripheral neuropathy (PN) (4,6,8). Although PET cannot directly display nerves, MRI combined with PET can help to clarify the anatomical location (6,8).

Our previous research confirmed the feasibility of using fluorine-18-fluorodeoxyglucose (¹⁸F-FDG) PET/MRI to diagnose PNI and evaluate its severity (2,9). They showed that PET uptake generally increases with severe nerve damage. However, considering that differences in the severity of damages among subjects could potentially influence the results, we deemed it necessary to evaluate varying degrees of nerve damage within the same subject.

The aim of the present study was to evaluate the diagnostic efficacy of ¹⁸F-FDG PET/MRI in the recovery process over time after PNI in the sciatic nerve of the same rat. To date, no research has been reported on consecutive PET scans performed during the recovery process after PNI in the same rat.

Material and Methods

Animals

Six male Sprague–Dawley rats (aged 7 weeks, weighing 200–250 g) were used in this study. They were individually housed in standard rat cages and maintained under constant temperature and humidity, with a 12-h light/dark cycle. They were provided with a pellet diet and purified drinking water and allowed to acclimate for one week before the experiment began. All experimental procedures in this study were approved by the Institutional Animal Care and Use Committee of Yonsei University Health System (Approval Number YUHS-2019-0293).

Experimental design

The experimental group comprised all six rats. The control group was replaced with one from the previous study (2). The surgical procedures were performed solely on the left sciatic nerves of the rats, while the right nerves were preserved. A 3-mm injury was induced on the freed left sciatic nerve proximal to the sciatic trifurcation using a curved hemostat (12.5 cm HB0515; HEBU, Germany) with a 40-N clamping force for 5 min. The degree of nerve damage was measured preoperatively and at one, two, three, four, and five weeks postoperatively using three different assessment methods: paw withdrawal threshold (PWT) test, PET, and MRI.

Surgical procedure

Under respiratory anesthesia with isoflurane, the left thigh was shaved and cleaned with 10% povidone-iodine. A skin incision was made 0.5 cm posterior to the femur. The fascia of biceps femoris and gluteus superficialis was exposed, and the sciatic nerve was approached by blunt dissection through the intermuscular space between them. The nerve was carefully freed from the surrounding tissue using micro-pincettes. Nerve injury was induced using the method described above. The surgical site was closed layer-by-layer using 4-0 Vicryl (Polyglactin 910; Ethicon, Inc., Somerville, NJ, USA) for the muscle and fascia and 5-0 Nylon (AILEE Co., Busan, Republic of Korea) for the skin. The animals were euthanized by CO₂ inhalation five weeks postoperatively.

Pain behavior assessment

Pain behavior assessments were conducted on the bilateral hind paws preoperatively and at one, two, three, four, and five weeks postoperatively. Each hind paw was measured three times, and the mean value was used. The PWT test was performed using the manual von Frey filament (BIO-VF-M; Bioseb Inc., Vitrolles, France), which is the gold standard for determining mechanical thresholds in rats and indicates force according to its size. The rats were placed in a customized modular holder cage for at least 15 min to acclimate. A monofilament was applied perpendicularly to the plantar surface of the hind paw until it bent, delivering a constant predetermined force for 2–5 s. The response was considered positive if the rats withdrew their paw during filament application. The plantar surface of the hind paw was the most commonly used area for testing, and the response was observed via a wire-gated floor. If a response was observed for any particular force, the rat was re-examined with a value one step below. If no response was noted, the particular value was recorded as the result. The researchers revised the withdrawal threshold values to compensate for individual rat conditions, such as sensitivity, nervousness, fear, adaptation, and blunting, before comparing variables at each PWT test. Revised withdrawal threshold (RevWT) values were calculated as follows:

R e v W T = \frac{Withdrawal thresholds value of the injured side (left) of each rat}{Average of withdrawal threshold values of the intact side (right) of rats}

Small-animal PET/MRI

The rats were fasted for 12 h before each PET scan. Animals underwent sequential small-animal PET (Inveon PET; Siemens, Germany) and MRI scans (Bruker 9.4 T 20-cm bore MRI system; Biospec 94/20 USR; Bruker, Ettlingen, Germany). To facilitate image superimposition, a customized PET table similar to the MRI table was used. An intravenous injection of approximately 1 mCi of ¹⁸F-FDG was administered. One-hour dynamic PET images of the thighs and T2-weighted rapid acquisition with relaxation enhancement MR images of the thighs (repetition time [TR] = 2300 ms; echo time [TE] = 11 ms; slice thickness = 1 mm; acquisition matrix = 192 × 192; acquisition field of view = 55 × 55 mm²) were obtained. The PET and MRI images were co-registered using the AMIDE image analysis software (amide.exe1.0.4; https://amide.sourceforge.net). MRI revealed the anatomic location of the sciatic nerves and the placement of the regions of interest (ROIs). To quantitatively analyze PET signals, spherical-shaped ROIs, measuring 6 mm in diameter, were placed around the nerve injury site and on the opposite side. The maximum standardized uptake values (SUVmax) of ROIs were calculated using OsiriX image analysis software (Pixmeo, Geneva, Switzerland) (Fig. 1a). The SUV ratio (SUVR) was calculated using the target and reference regions within the same PET image to eliminate differences between animals or points in time as follows:

S U V R = \frac{S U V_{t a r g e t}}{S U V_{r e f e r e n c e}} = \frac{S U V_{l e f t R O I}}{S U V_{r i g h t R O I}}

Fig. 1.

ROIs and sequential images of PET/MRI and MRI over time in Rat 2. (a) PET/MRI fused images after inducing left sciatic nerve injury. MRI, combined with PET, helped to clarify the location of the sciatic nerve on PET images. ROIs for PET analysis were set as a spherical-shape with 6 mm in diameter, and SUVmax values were measured in bilateral ROIs. The yellowish translucent circle indicates the left sciatic nerve injury area. The whitish translucent circle indicates the opposite area of left ROI, which includes an intact right sciatic nerve. The fluorine-18-fluorodeoxyglucose uptakes on the injured left sciatic nerve gradually decreased over time from 1 week to 5 weeks postoperatively (yellow dotted-line arrows). (b) MRI scans after inducing left sciatic nerve injury. The left sciatic nerve (ROI) revealed high signal intensity on the MRI scan after injury (yellow-line circle), and the right ROI (intact sciatic nerve) was drawn symmetrically (white-line circle). Additional ROI was positioned in the air (mint-line circle). Significant differences between the injured and intact sciatic nerves were not observed on the consecutive MR images except at 1 week postoperatively (yellow dotted-line arrows). MRI, magnetic resonance imaging; PET, positron emission tomography; ROI, region of interest.

For the quantitative analysis of MRI, the gray values (GV) of ROIs were measured on the axial sections of T2-weighted MRI scans using the ImageJ program (NIH, Bethesda, MD, USA). An additional ROI was set on an empty space (air) image to determine the contrast-to-noise ratio (10) (Fig. 1b). To compensate for variations among the animals, the ratio of relative signal intensity values (MRSIR) of ROIs was calculated as follows (10 –12):

M R S I R = \frac{G V_{t a r g e t R O I} - G V_{a i r}}{G V_{r e f e r e n c e R O I} - G V_{a i r}} = \frac{G V_{l e f t R O I} - G V_{a i r}}{G V_{r i g h t R O I} - G V_{a i r}}

Statistical analysis

The statistical analysis was conducted using the PASW SPSS software (version 18.0, SPSS Inc., Chicago, IL, USA). To confirm significant differences in parameters over time, the one-sample t-test was used for each assessment method. In addition, a linear regression analysis was performed to determine whether the three parameters were linearly related to elapsed time after the injury. Probabilities of ≤0.05 were considered significant.

Results

Mechanical threshold assessment

The PWT values of all rats at one and two weeks postoperatively were recorded as the same as the highest force among the manual von Frey filaments because the left paws of all rats did not respond to the thickest filament at those time points. Hypoesthesia was observed in the left legs of all rats only up to three weeks after injury. There was no significant difference in RevWT between before and >4 weeks after injury (P = 0.5) (Fig. 2). The results of the linear regression analysis indicate a significant decrease in RevWT over time (P = 0.000), with a significant explanatory power (adjusted R² = 0.89).

Fig. 2.

Results of mechanical threshold assessment. RevWT values of all rats significantly decreased over time after injury, and no significant difference was observed between before injury and 4 weeks after injury. At 1 and 2 weeks postoperatively, the paw withdrawal threshold values of all the rats were the same as the highest force among the manual von Frey filaments because the left paws of all the rats did not respond to the thickest filament at those time points.

Imaging analysis

PET

The PET analysis excluded the PET scan of rat 1 at five weeks postoperatively due to a technical problem causing a deformed image. Unlike other parameters, significant differences in SUVR were still observed between before and five weeks after the injury (P = 0.0007) (Figs. 1a and 3). In the linear regression analysis, SUVR significantly decreased over time (P = 0.000), and the explanatory power was significant (adjusted R² = 0.50).

Fig. 3.

Results of PET assessment. SUVR values significantly decreased over time after injury, and significant differences were observed between the values before injury and those at 5 weeks after injury. The PET scan of Rat 1 at 5 weeks postoperatively was deformed because of a technical problem, so it was excluded from the analysis. PET, positron emission tomography.

MRI

No significant difference in MRSIR was observed between before and >4 weeks after injury (P = 0.0627) (Figs. 1b and 4). The linear regression analysis showed a significant decrease in MRSIR over time (P = 0.000), with a significant explanatory power (adjusted R² = 0.37).

Fig. 4.

Results of magnetic resonance imaging assessment. MRSIR values significantly decreased over time after injury, and no significant difference was observed between the values before injury and those at 4 weeks after injury.

Discussion

This study is a follow-up to our previous research, which aimed to verify the usefulness of ¹⁸F-FDG PET/MRI in diagnosing PNI (2). Previous experiments have used different subjects with varying degrees of injuries to assess the differential diagnostic ability of assessment modalities for PNIs. However, this experiment was designed to evaluate the recovery process of damaged nerves within the same subject, because subjects may differ in sensitivity, degree of nerve damage, rate of recovery, response to the radioisotope, and the amount used. To date, there has been no research in which consecutive PET scans have been performed during the recovery process after PNI in the same subject. Furthermore, the evaluation methods commonly used for PN to date have inherent limitations, including subjectivity, invasiveness, susceptibility to environmental influence, and an inability to precisely determine the extent of nerve damage (2,4,13). Accurate diagnosis of impaired peripheral nerves is crucial for a positive prognosis. Recently, researchers have investigated and developed various non-invasive imaging modalities for PN, including MRI, US, and PET techniques. (4,13). These techniques show promise in addressing this issue.

MR technology is advancing rapidly. Advances in spatial resolution, processing, and available magnetic field strengths make it possible to evaluate the internal fascicular architecture of peripheral nerves using MRI. MR technology can also be used to evaluate peripheral nerves through MR neurography and diffusion-tensor imaging (DTI) (4,13,14). MR neurography, also known as high-resolution peripheral nerve MRI, is a technique that uses specific MR pulse sequences to enhance the visualization of peripheral nerves and distinguish them from surrounding structures (13 –15). This technique provides excellent soft tissue contrast, 3D images of entire nerves, and indirect information about the nerve's physiological status by evaluating the innervated muscles. DTI, which is derived from a diffusion-weighted imaging technique that captures the random microscopic motion of water molecules in tissues and generates images based on the difference in diffusion properties in various types of tissues, can be used to visualize the 3D course of nerves with tractography. As peripheral nerves are highly organized and preferentially allow water to diffuse along the nerve axis rather than perpendicular to its axis, anisotropy, which means the directional dependency of nerves, allows differentiation of nerve tissue from the surrounding isotropic tissues in DTI (4,13,14). While MR technology has advanced and has many merits, its use in evaluating peripheral nerves is limited due to inherent limitations, such as only static examination being available and hardware artifacts. Further research is necessary to apply DTI in vivo (13,14).

Technical advances in hardware and signal processing have allowed US to deliver high-resolution images in neuromuscular medicine and research (13,14). US has several advantages in evaluating nerves, including low cost, high resolution, dynamic capability to explore large areas of nerves, non-invasive procedure, no general contraindication, and widespread availability. However, the use of US for nerve assessment has significant limitations, such as the inability to visualize nerves without anatomical landmarks or nerves under bone, as well as limited depth of penetration (4,13,14). As a result, most studies on nerve assessment using US have relied on indirect methods to quantify affected muscles by neuropathy, and research into applying US to nerves directly is significantly less (13). A typical US does not provide sufficient information about the innervation status of a nerve, including Wallerian degeneration, regeneration, or axonal content, except for myelination abnormalities (13). While some researchers are exploring quantitative US neurography (13,16), it would be challenging to use US to directly assess peripheral nerves, particularly the inferior alveolar nerve, which passes through the mandible and is frequently damaged in dentistry.

Unlike the abovementioned imaging modalities, PET can promptly and sensitively detect differences in impaired nerves from normal ones, even if they appear intact, by visualizing their metabolism (2,4,8). Therefore, PET demonstrates high sensitivity in detecting pain-generating pathologies, while MRI and US focus solely on imaging anatomical alterations (6,17). Increased nociceptive activity is associated with increased metabolic, hemodynamic, mediator, and cellular changes. PET with specific molecular biomarkers may be applied to the molecular mechanisms, including cellular response, inflammatory mediators and receptors, ion channel expression, and metabolic response, to detect abnormal physiologic activity along the nociceptive pathway in the peripheral nervous system (2,6).

We performed a few animal studies consecutively using ¹⁸F-FDG PET/MRI to verify its diagnostic efficacy in PNI. ¹⁸F-FDG PET is familiar to researchers because it has been widely used for diagnosing malignant lesions. After nerve injury, increased neuronal activities depend on glucose metabolism and PET scan can demonstrate the regions of high metabolic activity through modified glucose molecules, such as ¹⁸F-FDG (4,6,8). Despite its promising potential in diagnosing PNs, few studies have used PET in this field, not in oncology (4,8,17,18). Therefore, we conducted research using ¹⁸F-FDG PET in this field and confirmed its potential as a valuable imaging modality for noninvasive and objective diagnosis of neuropathic pain caused by PNI in the previous studies (2,9).

This research compared two standard imaging modalities,¹⁸F-FDG PET and MRI, commonly used in clinical practice, to consecutively assess the recovery of injured peripheral nerve within the same rat. In addition, both imaging modalities were compared to a PWT test, which represents a clinical neurosensory test. The crushing injury model was selected for this study, based on previous research (2,9). The crushing-injured nerve maintained its continuity and was classified as axonotmesis by Seddon (19). There is no potential for localized inflammatory response to surgical sutures affecting PET imaging since it does not require threads to make with (2). To compensate for individual differences, we presented representative values for each assessment method by calculating the ratio of injured to intact sides for each rat. Each parameter decreased over time as the injured nerves recovered. However, a significant difference was still observed between each time point in the PET, unlike PWT and MRI, even at four and five weeks postoperatively. PWT and MRI indicated as if the nerve damage had been recovered during the same period. Although PWT had higher explanatory power than PET in the linear regression analysis, it is difficult to conclude that PWT is a better assessment method than PET due to the limitations of the manual PWT test in measuring at the early stages after damage in this experiment. Therefore, this experiment suggests that PET scanning can provide more detailed observation of the degree of recovery from nerve damage than the other two assessment methods.

The present study has some limitations. The number of experimental animals and the duration of the experiment were limited due to the restricted production and use of radioisotopes. In addition, we were unable to obtain tissue samples for histomorphometric analysis at each recovery phase, which prevented a comparison of imaging and histology results. To learn about tissue sample findings, please refer to our previous research (2,9). Experimental validation of the clinical application of PET/MRI in PNIs may not be feasible in large animals due to equipment limitations. Therefore, it is expected that well-designed studies with a large number of animals will provide a reliable basis for clinical trials. Additionally, further research should be conducted to develop more reliable assessment methods for PN, utilizing molecular biomarkers and contrast agents that are currently under development.

In conclusion, conventional methods for assessing PN may not always align with a patient's clinical presentation, making diagnosis challenging. This study suggests that PET and MRI could be valuable non-invasive techniques for diagnosing neuropathic pain resulting from PNI. PET/MRI would be expected to be a more accurate and informative diagnostic tool for PNI than MRI alone.

Footnotes

Acknowledgements

The authors thank Professor Hoi In Jung of the Department of Preventive Dentistry and Public Oral Health, Yonsei University College of Dentistry for her strong support in this study.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by Wonkwang University (2024).

ORCID iD

Jung Woo Nam

References

Meacham

Shepherd

Mohapatra

, et al. Neuropathic pain: central vs. Peripheral mechanisms. Curr Pain Headache Rep 2017;21:28.

Nam

Lee

Kim

. Diagnostic efficacy of (18)F-FDG PET/MRI in peripheral nerve injury models. Neurochem Res 2019;44:2092–2102.

Ahn

Park

. Animal models for orofacial neuropathic pain. Hanyang Med Rev 2011;31:107–115.

Rangavajla

Mokarram

Masoodzadehgan

, et al. Noninvasive imaging of peripheral nerves. Cells Tissues Organs 2014;200:69–77.

Technology review: the Neurometer Current Perception Threshold (CPT). AAEM Equipment and Computer Committee. American Association of Electrodiagnostic Medicine. Muscle Nerve 1999;22:523–531.

Tung

Behera

Biswal

. Neuropathic pain mechanisms and imaging. Semin Musculoskelet Radiol 2015;19:103–111.

Ohana

Moser

Moussaoui

, et al. Current and future imaging of the peripheral nervous system. Diagn Interv Imaging 2014;95:17–26.

Behera

Jacobs

Behera

, et al. (18)F-FDG PET/MRI can be used to identify injured peripheral nerves in a model of neuropathic pain. J Nucl Med 2011;52:1308–1312.

Park

Lee

Kim

, et al. Evaluation of peripheral nerve injury according to the severity of damage using (18)F-FDG PET/MRI in a rat model of sciatic nerve injury. Neurol Res 2024;46:356–366.

10.

Marquez Neto

Freitas

Mendelovitz

, et al. An initial clinical experience to improve postoperative monitoring of peripheral nerve regeneration following neurotmesis using magnetic resonance imaging at 1.5 Tesla. J Neurosurg Sci 2016;60:329–338.

11.

Behr

Schnabel

Mirastschijski

, et al. Magnetic resonance imaging monitoring of peripheral nerve regeneration following neurotmesis at 4.7 Tesla. Plast Reconstr Surg 2009;123:1778–1788.

12.

Matsuda

Wang

Suo

, et al. Retrograde axonal tracing using manganese enhanced magnetic resonance imaging. Neuroimage 2010;50:366–374.

13.

Mandeville

Deshmukh

Tan

, et al. A scoping review of current and emerging techniques for evaluation of peripheral nerve health, degeneration and regeneration: part 2, non-invasive imaging. J Neural Eng 2023;20:1–25. DOI: https://doi.org/10.1088/1741-2552/ace217.

14.

Holzgrefe

Wagner

Singer

, et al. Imaging of the peripheral nerve: concepts and future direction of magnetic resonance neurography and ultrasound. J Hand Surg Am 2019;44:1066–1079.

15.

Sneag

Queler

. Technological advancements in magnetic resonance neurography. Curr Neurol Neurosci Rep 2019;19:75.

16.

Jerban

Barrere

Andre

, et al. Quantitative ultrasound techniques used for peripheral nerve assessment. Diagnostics (Basel) 2023;13(956):1–26.

17.

Shen

Behera

James

, et al. Visualizing nerve injury in a neuropathic pain model with [(18)F]FTC-146 PET/MRI. Theranostics 2017;7:2794–2805.

18.

Lasko

Huang

Voorbach

, et al. Multimodal assessment of nervous and immune system responses following sciatic nerve injury. Pain 2013;154:2782–2793.

19.

Seddon

. Three types of nerve injury. Brain 1943;66:237–288.