Processed nerve allografts to repair peripheral nerve discontinuities

4.1 In a randomised controlled trial (RCT) of 23 patients needing digital nerve repair comparing processed nerve allograft (PNA) with treated bovine graft at 12‑month follow-up, static 2‑point discrimination assessment (s2PD, which tests the ability to discern the difference between 1 and 2 static pressure points) was statistically significantly better in the PNA group (n=5) than the bovine graft group (n=7; 5±1 mm versus 8±5 mm, p<0.05). In the same study, moving 2‑point discrimination assessment (m2PD) was not statistically significantly different between the PNA group and the bovine graft group (5±1 mm versus 7±5 mm, p>0.05) at 12‑month follow-up. In a non-randomised comparative study of 153 patients needing digital nerve repair comparing PNA repair (n=72) with tension-free suture nerve repair (n=81), s2PD scores (excellent plus good, defined as the ability to distinguish between 2 static pressure points at a maximum distance of 15 mm) were not statistically significantly different between the PNA group (67% [48/72]) and the tension-free suture group (64% [52/81]) at 6‑month follow-up (p=0.749). In a case series of 17 patients with digital nerve injuries treated by PNA grafting, s2PD was excellent or good in 78% (14/18) of digits repaired, at a mean follow-up of 15 months. In the RCT of 23 patients, Semmes–Weinstein monofilament test (testing of pressure threshold using a monofilament; range: 2.833=normal sensation to 6.650=residual sensation) was statistically significantly better in the PNA group than the treated bovine graft group (3.6±0.7 versus 4.4±1.4, p<0.05) at 12‑month follow-up. In the same study, thermal sensation was totally improved from baseline at 12‑month follow-up and not statistically significantly different between the treatment (PNA group: from 7% [1/14] to 100% [6/6] and bovine graft group: from 33% [3/9] to 100% [7/7]).

4.2 In a case series of 64 patients needing nerve repair in the upper extremity and treated by grafting using PNA, there was meaningful recovery in 75% (48/64) of all patients. Univariate analysis showed that distal sites of injuries have a statistically significantly higher likelihood of recovery than proximal upper limb sites (odds ratio [OR] 5.606, 95% confidence interval [CI] 1.663 to 18.903; p<0.05). In the same study, discontinuities smaller than 30 mm had a statistically significantly greater likelihood of meaningful repair than those greater than 50 mm (OR 14.333, 95% CI 2.143 to 95.848; p<0.05). In a case series of 26 patients with lingual nerve and inferior alveolar nerve discontinuities treated by PNA grafting, meaningful sensory recovery was assessed using a neurosensory test improvement tool (ranging from normal=best, through mild, moderate and severe to complete=worse). At 12‑month follow-up, neurosensory test improvement scores were normal in 52% (12/23), mild in 9% (2/23), moderate in 26% (6/23) and severe in 13% (3/23) of patients. In the same study, neurosensory improvement was reported in 86% (12/14) of patients with discontinuities 8 mm to 20 mm in length and 89% (8/9) of patients with discontinuities 30 mm to 70 mm in length.

4.3 In the RCT of 23 patients, disability of the arm, shoulder and hand score (DASH: 0=no disability, 100=most severe disability) was not statistically significantly different between the PNA group (5±6.5) and the bovine graft group (8±6.3) at 12‑month follow-up (p=0.318).

4.4 In a case series of 108 patients needing nerve repair, there was no sensory recovery because of graft failure in 5% (4/76) of patients at last follow-up and surgical revision was needed.

4.5 In the RCT of 23 patients, at 12‑month follow-up, pain measured using a visual analogue scale (VAS, 0=no pain, 10=extreme pain) had improved from baseline in both groups (PNA group: from 4.7±3.4 to 0.5±0.6; treated bovine graft: from 4.4±2.1 to 0.9±1.0) but there was no statistically significant difference between the groups (p=0.432). In another case series of 26 patients needing PNA after resection of neuromas of the foot and ankle, mean ordinal pain score (0=no pain to 10=worse pain) statistically significantly reduced from 7.5 points at baseline to 4.9 points at a mean 66‑week follow-up (difference 2.6, range +2.0 to -8.0; p=0.016). In the same study, patient reported outcome measurement information system scores were used to assess the impact of pain on patients' behaviour and daily function (reported as T‑scores with a population mean of 50 and a standard deviation of 10). Pain behaviour T‑score decreased by 7.3 (range+2.0 to -22.0) from 63.0 at baseline (percentile decrease of 24%, p<0.003). Pain interference T‑score decreased by 11.3 (range +2.0 to -27.0) from 68.0 at baseline (mean percentile change of 31%, p<0.003). In a case series of 17 patients with digital nerve injury treated by grafting with PNA, pain (measured using a VAS: 0=no pain, 10=extreme pain) worsened in 1 patient (VAS score increased from 5 at baseline to 8 at 15‑month follow).

4.6 In the non-randomised comparative study of 153 patients, difference in satisfaction rate was not statistically significantly different between the PNA group and the tension-free suture group (2.02%, 95% CI -6.07 to 10.87) at 6‑month follow-up.

4.7 The specialist advisers listed key efficacy outcomes as re-innervation of target organs, nerve regeneration rate, clinical sensory and motor outcome scales, and patient reported outcomes.