The outcome of polyethylene glycol fusion augmented by electrical stimulation in a delayed setting of nerve repair following neurotmesis in a rat model

The animals were given a gavage of non-steroidal anti-inflammatory drug, carprofen (5 mg/kg /OD), 1 h before surgery. They were divided into four groups, each consisting of six animals. Group I represented the controls undergoing routine nerve transection and traditional epineural repair under the microscope immediately. Group II underwent nerve transection, repair, and PEG fusion immediately. Group III underwent nerve transection, but repair and PEG fusion were done after 48 h. Group IV underwent the same as group III but was additionally subjected to per-operative electrical stimulation of 20 Htz for 1 h proximal to the site of nerve fusion.

The animals were anesthetized with an intraperitoneal injection of a cocktail containing xylazine (10 mg/kg) and ketamine (75 mg/kg). Under aseptic conditions, the sciatic nerve was exposed. In groups I and II, the sciatic nerve was transected midway between the sciatic notch and its distal trifurcation. A primary epineural neurorrhaphy with two sutures was done under a microscope with 10–0 nylon sutures (day 0) in group I (controls). In group II, PEG fusion was done as follows: The nerve endings, after transection, were irrigated with a hypotonic solution, Plasmalyte-A™ (Baxter) (Fig. 1A) [7, 11]. This removes the Ca²⁺ ions which are known to seal and close the axon ends immediately after the nerve is cut by forming intracellular vesicles [7, 11]. The hypotonic saline would expand the open axonal ends, which could then be prepared for apposition by PEG-fusion to the distal cut end [6, 9]. This is immediately followed by the application of 1% methylene blue in sterile water at both nerve ends, which is an antioxidant and maintains the Ca²⁺ free environment and prevents additional vesicle formation (Fig. 1B) [7, 11]. Primary neurorrhaphy is then done by the same technique as for group I. It provides mechanical strength at the repair site and prevents PEG-fused axons from pulling apart as the plasmalemma has a low tensile strength (Fig. 1C) [7, 11]. We then applied a solution of 50% PEG (Sigma-Aldrich, 3.35 kD) to the site of repair for a minute (Fig. 1D) [6, 10]. The wound was rinsed with the calcium-containing Ringer’s lactate solution, to remove the residual PEG solution and provide Ca²⁺ to repair axolemmal holes with vesicles (Fig. 1E) [7, 11]. In groups III and IV, the nerve was crush ligated with 6–0 nylon sutures to achieve neurotmesis at the same site as described for groups I and II. After 48 h, the ligature was reexplored and divided at the ligation site and the segment was examined under the microscope to confirm damage to all the fascicles. The ends were freshened with microscissors followed by neurorrhaphy and PEG fusion in the same way as done for group II. In group IV, we additionally gave ES of 20Htz at 2 mA using a manual electrical stimulator (Indmed™) for 1 h immediately following the repair at a site 5 mm proximal to the repair site (Fig. 1F) [2, 9, 11, 20]. The wound was closed with 4–0 nylon sutures followed by subcutaneous infiltration of gentamycin (2 mg/kg.). The animals were returned to the housing facility and observed for complications like wound dehiscence and chest or gastrointestinal infections. Carprofen was given orally as an analgesic for 5 days.

The pinprick test was done to assess the sensory recovery in the feet (Fig. 2D). The unoperated foot was tested with a 25-G needle pinprick stimulus to document the normal reaction, represented by an immediate foot withdrawal, and compared with the operated side. The foot was pricked on the plantar side of the lateral half. The responses were graded as suggested by Paskal et al. [17] ( −) no response; ( +) mild response, a very weak retraction or trembling of the leg; (+ +) moderate response; and (+ + +) normal response, as observed on the healthy side [17]

At 3 weeks post-operatively, animals were tested for cold allodynia response, as suggested by Jo et al. (Fig. 2C) [13]. The animal was placed on a meshed metal grid for 5 min and allowed to familiarize itself with the setup. A drop of acetone was placed using a syringe without its needle on the plantar side of the operated foot. The animal’s reaction was observed for a minute [14]. Generally, there is no response or a minimal reaction to cold acetone applied to the foot pad of normal rats [14]. Similarly, no response will be seen with absent sensations. However, after nerve injury, typical behavioral responses can be observed with partial recovery on applying cold acetone, indicating increased cold sensitivity. A recovering or partially recovered nerve will show paw-licking/shaking behavior. This is accepted as a positive response to acetone suggesting sensory recovery [14]. The total time of behavior exhibited by the animal was also measured.

On the 21st day, all the animals were euthanized, and the sciatic nerve samples of a length of 1 cm immediately distal to the site of repair were collected for hematoxylin and eosin-based histological study for axonal counts. The legs’ gastrosoleus and extensor compartment muscles (tibialis anterior and extensor digitorum longus and hallucis longus) were checked for their weights using a digital weighing scale. The affected to contralateral side muscle weight ratio was documented and recorded as relative muscle mass.

The sciatic nerve sample was fixed in 4% paraformaldehyde in phosphate-buffered saline (PBS) for 3 h [9]. Before embedding, the nerve sample was washed and preserved in 0.2 g glycine in 100 mL PBS. It was dehydrated serially with 50, 70, 80, 95, and 100% ethanol passages and embedded in paraffin [9]. A 7- to 8-mm thick series of transverse sections were cut using a Leica™ microtome, placed on slides, and put for overnight drying. They were then deparaffinated and rehydrated with ethanol in decreasing strength [9]. The slide was immersed in 0.1% hematoxylin for 10 min and washed in water for 15 min. It was then washed in distilled water after immersion in 0.1% eosin for 5 min. The slides were then analyzed at 10 × and 50 × on an Olympus microscope. The open-source ImageJ (NIH, Bethesda) software was used for nerve fiber counts. The muscle sample was sent for weight measurement analysis using a weighing scale. The analysis was done by an observer blinded to the four groups.

All statistical analyses were performed by Jamovi software. Data was entered as mean ± SD. Data for SFI was assessed with a two-way analysis of variance (ANOVA). The differences between groups in the axonal count and the functional test results were verified with an unpaired T-test. The threshold of statistical significance was set at p ≤ 0.05.

As per Table 1, the pre-operative SFI was uniformly normal and equally distributed in all four groups. One week after the surgery on the affected side, high values of SFI confirm the neurotmesis in all the groups, suggesting a lack of motor recovery across all the groups. The second-week values show a recovery pattern that is significantly better (p < 0.001) in all the intervention groups (Table 1) when compared with the control. In the third week, all three intervention groups showed significantly better recovery (p < 0.001) in comparison to control (Table 1). The results showed a better recovery pattern for the experimental groups when compared to the control (Group I) (Fig. 2A and B). This suggests that a good volume of axons may be available for PEG fusion at 48 h after neurotmesis. Group II showed the most improved results, followed closely by group IV. Tukey post hoc tests at 3rd week between groups III and IV suggest a significantly better motor recovery (p < 0.001) in group IV. We observed that using electrical stimulation helped improve results in the delayed repair at 48 h.

The skin prick test result was considered a measure of the sensory function return. One week after the surgery, none of the study animals responded to the foot prick stimulus. Pinprick grades were higher in the experimental groups than in the control group, but the difference was not statistically significant among the experimental groups (p = 0.01) (Table 2) (Fig. 3). The results suggest that partial recovery of sensation does occur even by 3 weeks which shows an overall higher value in the experimental groups. The recovery in groups III and IV is as good as in the acute PEG repair group (p = 0.01).

Groups II, III, and IV showed a response to cold stimuli. The cold allodynia response was absent in the control group, suggesting that sensation recovery had not happened by 3 weeks. In the unpaired T-test, a comparison of groups III and IV demonstrates that a significant early response was seen in group IV (p < 0.001) (Fig. 4). The intervention groups showed a trend toward faster recovery of sensation in comparison to the control group. The difference in the cold allodynia response was insignificant between groups II and IV (p = 0.01), suggesting a similar response of sensory recovery. This shows the possible benefit of adding ES to the regime of PEG fusion in delayed repairs.

The relative muscle mass ratio for both the gastrocnemius(plantar flexors) and ankle and toe extensor muscles (dorsiflexors) remained lower than the contralateral normal side, suggesting that Wallerian degeneration cannot be prevented entirely even after acute PEG fusion (Table 2). However, the value of all the experimental group animals was more than the control. In the unpaired T-test, a comparison of groups III and IV demonstrates a significantly higher weight of plantar flexors in group IV (p < 0.001) (Fig. 5). Similarly, in the unpaired T-test for dorsiflexors, a comparison of groups III and IV demonstrates a higher weight of dorsiflexors in group IV (p < 0.05) (Fig. 4). The difference was better reflected for the plantar flexors among the intervention group when compared to the control. An important reason which could explain this is that the nerve to the gastrosoleus is closer to the site of repair and may have recovered faster to show a marginal improvement in its volume.

In this series, total nerve fiber counts (the number of axons) increased distal to the repair site in all three groups treated with PEG compared with the repair-alone control group (Table 2). One-way ANOVA results showed a significant difference between the control (group I) and the other groups for the axon count during 3rd week (F = 117.82, p < 0.001) (Table 1) (Figs. 6 and 7). In the unpaired T-Test comparison of groups III and IV, there is a mild increase in axon count in group IV, and no statistical significance is seen.

When the efficacy of interventions applied for delayed settings is compared between groups III and IV, group IV demonstrates significant therapeutic efficacy, as seen for SFI, dorsiflexors, and plantar flexors. Early response is seen for cold allodynia tests underlining the benefits of ES.