Available online at

ScienceDirect www.sciencedirect.com Chirurgie de la main 33 (2014) 279–285

Original article

Sensory recovery after primary repair of palmar digital nerves using a Revolnerv1 collagen conduit: A prospective series of 27 cases Étude de la récupération sensitive après réparation primaire des nerfs digitaux palmaires à l’aide d’un conduit en collagène Revolnerv1 : à propos d’une série prospective de 27 cas A. Arnaout *, C. Fontaine, C. Chantelot Service orthopédie B, hôpital Roger-Salengro, centre FESUM, « SOS-mains Lesquin », centre hospitalier régional universitaire de Lille, rue du Pr.-Émile-Laine, 59037 Lille cedex, France Received 31 October 2013; received in revised form 18 April 2014; accepted 19 May 2014 Available online 18 June 2014

Abstract Despite advances in microsurgery, digital nerve repair remains a challenge due to the lack of reproducible procedures with satisfactory functional results. The aim of this study was to compare the sensory and functional results of direct microsurgical sutures protected by a Revolnerv1 nerve regeneration conduit, with results of a series of direct sutures without a protective conduit in the literature. From November 2009 to April 2010, 35 patients were treated by direct epiperineural suture for digital nerve injury, protected by a Revolnerv1 nerve regeneration conduit at the FESUM centre ‘‘SOS-mains Lesquin/CHRU de Lille’’. Sensory recovery was assessed by the static two-point discrimination Weber test (WS) and the Semmes-Weinstein (SW) test at postoperative months 1, 3, and 6. The final evaluation was performed after a minimum follow-up of 6 months. Statistical analysis of sensory results (WS and SW) was mainly performed with non-parametric tests (Wilcoxon, Mann and Whitney). P < 0.05 was considered to be statistically significant. One patient was excluded, six were lost to follow-up, and four could not be seen at the 6month follow-up visit. Finally, 24 patients and 27 nerve sutures were included. Mean age was 38 years old and the ratio of women/men was 1/5. Eighty-five percent of the patients had useful (S3 + ) or normal (S4) discrimination at 6 months, and the average WS was 10.3 ( 3.76). There was a tendency to better WS results in sharp transections compared to jagged lacerations (9.19 vs 11.82). The SW test was satisfactory in 15% of patients and acceptable in 30%. There were no complications from the Revolnerv1 collagen tube. After 6 months follow-up this study shows that results with the Revolnerv1 nerve regeneration conduit on direct palmar digital nerve sutures were comparable to but not better than those of uncoated direct sutures. A study including a larger population with longer follow-up is necessary to determine the value of this technique and its recommendation for general use in all digital nerve injuries. # 2014 Elsevier Masson SAS. All rights reserved. Keywords: Digital nerve; Digital lesion; Nervous regeneration chamber; Nerve conduit; Nerve tube

Résumé La réparation chirurgicale des lésions nerveuses digitales demeure un problème malgré les avancées de la microchirurgie en raison de l’absence de procédé reproductible donnant des résultats fonctionnels satisfaisants. L’objectif de ce travail prospectif était de faire le bilan sensitif et fonctionnel des sutures microchirurgicales directes enrobées par un tuteur Revolnerv1, et de comparer ces résultats à ceux des séries de sutures directes sans enrobement de la littérature. De novembre 2009 à avril 2010, 35 patients ont bénéficié d’une suture d’une lésion nerveuse digitale protégée par un tube nerveux Revolnerv1 dans le centre FESUM « SOS-main Lesquin/CHRU de Lille ». L’évaluation des résultats fonctionnels sensitifs reposait sur le test de Weber statique (WS), le test aux monofilaments de Semmes-Weinstein (SW), réalisés aux 1er, 3e, 6e mois. Un recul minimal de 6 mois était requis pour l’évaluation finale. L’analyse statistique des résultats sensitifs (WS et SW) était réalisée principalement avec

* Corresponding author. E-mail addresses: [email protected] (A. Arnaout), [email protected] (C. Fontaine), [email protected] (C. Chantelot). http://dx.doi.org/10.1016/j.main.2014.05.002 1297-3203/# 2014 Elsevier Masson SAS. All rights reserved.

A. Arnaout et al. / Chirurgie de la main 33 (2014) 279–285

280

des tests non paramétriques (Wilcoxon, Mann et Whitney). Une valeur de p < 0,05 était considérée comme statistiquement significative. Un patient a été exclu, six autres ont été perdus de vue et quatre n’ont pas pu être revus à 6 mois. Au total, 24 patients ont été inclus et 27 sutures nerveuses ont été évaluées. L’âge moyen était de 38 ans (13,5–71) et le sex-ratio femme/homme de 1 vs 5. Quatre-vingt-cinq pour cent des patients avaient récupéré une discrimination utile (S3 + ) ou normale (S4) à 6 mois, et le WS moyen était de 10,3 ( 3,76). Seule une tendance à un WS plus favorable pour les sections nettes était observée comparativement aux lésions contuses (9,19 vs 11,82; p = 0,06). Le test de SW était satisfaisant chez 15 % des patients et acceptable chez 30 % des patients. Aucune complication imputable au Revolnerv1 n’a été observée. Les résultats à 6 mois de l’utilisation du conduit Revolnerv1 dans le cadre des sutures directes des lésions de nerfs digitaux palmaires sont, dans notre étude, comparables à et pas meilleurs à 6 mois que ceux de sutures directes non enrobées de la littérature. Un nombre plus important de patients avec un recul supérieur est nécessaire avant de conclure sur la supériorité de cette technique et sur l’intérêt réel de la généraliser à toutes les lésions nerveuses digitales. # 2014 Elsevier Masson SAS. Tous droits réservés. Mots clés : Nerf digital ; Lésion nerveuse ; Chambre de régénération nerveuse ; Conduit nerveux ; Tube nerveux

1. Introduction

2. Patients and methods

Because of their superficial anatomical location, palmar digital nerves are the most frequently injured of the peripheral nerves. Despite advances in microsurgery, surgical repair of these lesions is still a challenge due to the absence of reproducible procedures with satisfactory functional results based on international classifications. Indeed, although direct nerve suture [1,2] is still the generally accepted gold standard, overall results are poor [1] (one third good, average and poor results, respectively) (Table 1). The aim of this study was to evaluate the results of the collagen nerve regeneration conduit Revolnerv1 which was developed to create optimal conditions [3] for nerve regeneration in cases of emergency digital nerve repair, and to assess the value of this technique based on objective sensory (SW and WS tests) and functional criteria (international classification).

2.1. Study population

Table 1 Results of primary epineural sutures. Number of cases S2 (%) S3 (%) S3+ (%) S4 (%) Bunnel, 1927 Larsen, 1958 Weckesser, 1961 Weckesser, 1961 Onne, 1962 Onne, 1962 Honner, 1970 Honner, 1970 Buncke, 1972 Poppen, 1979 Posh, 1980 Young, 1981 Sullivan, 1985 Altissimi, 1991 Chaise, 1993 Total Average Standard deviation Minimum Maximum

105 142 12 12 8 14 24 50 18 62 71 27 42 54 110 731

? 7 0 16 0 28 12 28 0 10 20 0 26 26 18

? 29 16 16 0 29 20 28 22 16 32 10 28.5 24 14

? 64 42 21 0 43 20 28 28 55 48 57 28.5 37 48

? 0 25 63 100 0 48 16 50 19 0 33 17 13 20

13.64 11.04 0 28

20.32 8.96 0 32

37.10 17.32 0 64

28.85 28.15 0 100

From Chaise F. Plaies des nerfs collatéraux palmaires des doigts. Cahiers d’enseignement de la SOFCOT, 1997, 64 : p 110 [2].

All patients presenting with a transected (common or proper) palmar digital nerve who underwent emergency surgery, treated by direct epineural nerve suture associated with a Revolnerv1 nerve regeneration conduit were included in the study. Exclusion criteria were the following: the presence of an active infection, wounds with a high risk of infection, dorsal digital nerve wounds, complete amputation of the finger, a skin graft or a flap of the finger tip. Regular, continuous follow-up was performed from November 2009 to April 2010. Eight hand surgeons participated in the study. 2.2. Revolnerv1 collagen conduit (Orthomed SA06640 Saint-Jeannet, France) This nerve regeneration conduit made of type I and III porcine collagen has CE approval allowing clinical use in humans. It meets to the technical specifications of a nerve regeneration conduit [4–6]. The tubular shape physically guides axon regeneration in the conduit, and results in a concentration of neurotropic and neurotrophic factors at the suture site. The strong, stable tube wall provides mechanical protection of the suture zone from the surrounding environment and limits adhesions by fibrous scar tissue. Collagen, which is a major component of the extracellular matrix, favours the migration of Schwann cells [7]. 2.3. Surgical technique The choice of the surgical approach to expose the nerve was left up to the surgeon, and was based on local conditions and associated injuries (Fig. 1). After cleaning dissection of the nerve stumps and cutting the damaged nerve ends, an initial epiperineural suture was performed with Ethilon1 8.0 or 9.0. The surgeon chose the size of the conduit that seemed best adapted to the damaged nerve. The following sizes (diameter  length) were available (in mm): 2  10, 2  20, 2  15 and 4  10. The conduit was implanted as follows: the tube was opened longitudinally, then the pre-hydrated tube was

A. Arnaout et al. / Chirurgie de la main 33 (2014) 279–285

281

Sensory recovery was evaluated by the Weber test (WS) with a static two-point discrimination tester (in mm) at each followup consultation and the Semmes-Weinstein monofilament test (SW) (simplified kit with 5 monofilaments). The results of the WS were interpreted according to modified ASSH guidelines [8,9] (Table 2), and the SW test according to the IMAI classification [10] (Table 3). Finally, sensory results were stratified into different grades of functional recovery according to the British Medical Council classification [1,11] (Table 4). Clinical sensory tests were performed at each follow-up visit and compared to the contralateral hand and the corresponding hemipulp [11]. 2.5. Statistical analysis Because of the number of patients in this study, only nonparametric tests could be performed in the different subgroups (Wilcoxon, Mann and Whitney; SAS software version 9.2). P < 0.05 was considered to be statistically significant. 3. Results 3.1. Clinical series

Fig. 1. Surgical technique.

From November 2009 to April 2010, 35 patients were treated with this technique. Four were not included in the final analysis, because there was less than 6 months follow-up. One patient was excluded (secondary amputation of a crushed finger). Six patients were lost to follow-up. The final analysis included 24 patients and 27 nerve injuries repaired by direct epiperineural suture and the Revolnerv1 conduit. The cohort included 4 women and 20 men (gender ratio: 1 to 5), mean age 38 years old (13.5–71 years old), and the peak incidence was 32 years old. Emergency surgery was performed in all patients less than 24 hours after injury. Primary repair was performed in all cases. The dominant side was injured in 10/27 cases. There were sharp transections in 16/27 cases and jagged lacerations

rolled around the sutured area and the surgeon confirmed the stability of the tube thus excluding any risk of spontaneous expulsion.

Table 2 Modified ASSH guidelines [8,9].

2.4. Data collection Perioperative and postoperative assessments were performed. The following information was obtained about the patient during surgery (age, dominant hand, professional activity, tobacco use, diabetes), the injury (mechanism of injury, site of injury, associated injuries). The postoperative assessment was performed by one examiner (AA), during the three follow-up visits at postoperative months 1, 3 and 6. The goal of the first visit was mainly to identify any early complications due to the Revolnerv1 including allergic reactions, intolerance or rejection. Complications (abnormal bleeding, signs of infection, pain, complex regional pain syndrome, problems with healing of the incision) were also looked for. Later visits focused on an evaluation of sensory and functional recovery.

Results of the static WEBER test (mm)

Clinical functional correlation

< 6 mm 6–10 mm 11–15 mm > 15 mm

Excellent Good Fair Poor

Table 3 IMAI classification [10]. SW test 2.83 3.61 4.31 4.56 6.65 6.65

(green) (blue) (purple) (red) (red) recognized (red) not recognized

Clinical results Normal tactile sensation Diminished light touch Diminished protective sensation Loss of protective sensation Basic deep pressure Insensitive

A. Arnaout et al. / Chirurgie de la main 33 (2014) 279–285

282

Table 4 British Medical Council classification [1,11] and results of the Revolnerv1 series. Score

Static WEBER test

Corresponding function

Revolnerv1 (n = 27)

S0: no recovery S1: pain S2: tactile + hyperesthesia S3: tactile w/o hyperesthesia S 3+: useful discrimination S4: normal

– – – 15–40 mm 7–15 mm  6 mm

– Protection Protection Basic grasping Object recognition Normal

0 0 0 4/27 (15%) 17/27 (63%) 6/27 (22%)

(crushing or a circular saw) in 11/27 cases. One associated injury was present in 11/27 cases (fracture, tendon injury, loss of skin substance). Associated injury of the corresponding palmar digital artery was present in 10/27 cases (37%). According to the surgical report, three of these arteries were sutured. The accident was work-related in 4/24 patients. There was no abnormal local inflammation, or severe scar complications due to the Revolnerv1 conduit, which might have suggested an allergic reaction, intolerance or rejection.

3.3. Results of the Semmes and Weinstein test The results of the SW test are reported in Fig. 2. At the 6-month postoperative follow-up the results were satisfactory in 15% of the patients (sensation or diminished light touch), nearly one third of the patients had acceptable functional results (diminished protective sensations), and 55% had mediocre functional results (loss of protective sensation and pulp insensitivity), when the SW test was considered alone.

3.2. Weber two-point discrimination test The overall mean WS was 10.3 (SD = 3.76; median = 11.00). Most (nearly 3/4) of the results tended to be good or fair. Good and excellent results were obtained in 41%. The WS seemed to be significantly better in the group with sharp transections than in those with jagged lacerations (good or excellent results in 56% compared to 18%, respectively and a WS that was 3 points lower) (Table 5). Nevertheless, no statistically significant difference was found in WS results between sharp transections and jagged lacerations, there was only a tendency towards better results in the former (Wilcoxon test, P = 0.0696). The mean WS in the group with palmar artery injury was 11.60 (SD: 2.01; median: 11.00) compared to 9.5 (SD: 4.36; median: 7.00) in the group without. There were good or excellent results in 59% of the group without artery injury (10% for the other group) (Table 6). However, there was no statistically significant difference between the two subgroups.

Fig. 2. Global results of monofilament tests in the present series.

Table 5 Results of the static two-point discrimination Weber test (WS) for the Revolnerv1 series. Comparison with the Altissimi and Efstathopoulos series.

Revolnerv1 (n = 27) Revolnerv1 sharp transection (n = 16) Revolnerv1 jagged lacerations (n = 11) Altissimi (n = 54) Efstathopoulos (n = 64)

Excellent < 6 mm

Good 6–10 mm

Fair 11–15 mm

Poor > 15 mm

11% (3/27) 19% (3/16) 0% (2/11) 13% 14%

30% (8/27) 37,5% (6/16) 18% (2/11) 37% 34%

44% (12/27) 31% (5/16) 64% (7/11) 24% 23%

15% (4/27) 12,5% (2/16) 18% (2/11) 26% 28%

Table 6 Results of the static two-point discrimination test (Weber test) in relation to palmar digital artery injury.

Palmar proper digital artery+ Palmar proper digital artery–

Excellent

Good

Fair

Poor

0% 18% (3/10)

10% (1/10) 41% (7/17)

80% (8/10) 23% (4/17)

10% (1/10) 18% (3/17)

A. Arnaout et al. / Chirurgie de la main 33 (2014) 279–285

283

Fig. 3. Results of the monofilament test according to mechanism of injury.

Distribution of the monofilament test into the 5 categories was not significantly different for the subgroups ‘‘sharp transection’’ and ‘‘jagged laceration’’ (Fig. 3). At the 6-month follow-up, most patients had recovered normal sensitivity or useful discrimination allowing object recognition according to the international classification (S3+ and S4 = 85%) (Table 4). 4. Discussion 4.1. The global WS test According to the modified ASSH guidelines, the results of the WS test in our study were good or excellent in 41% and fair or mediocre in 59%. Although the 6-month postoperative follow-up was short, this is the minimum follow-up necessary to obtain stable, interpretable results [12]. The high rate of fair and mediocre results (59%) can be explained by the significant number of patients who smoked (nearly 50%), and of jagged lacerations (41%) as well as the relatively short follow-up. Because of the size of this cohort, only non-parametric tests could be used to evaluate the WS test in the different subgroups. In relation to the mechanism of injury, we only found a tendency for better results in the group with a sharp transection. The results in the literature also show better results with this type of injury [2,9,12–15]. In our study, the lack of relationship could be due to the lack of statistical power because of the limited number of patients in each subgroup. In relation to functional results, most patients recovered useful discrimination or normal function (85% classified S3+ et S4). Monofilament tests showed complete or nearly complete recovery in 15% of cases, satisfactory recovery in 30% and mediocre in 55% of cases. These results are disappointing compared to the WS test. Nevertheless in 6 of these nerves, the contralateral hand was compared and ‘‘light touch’’ was found to be diminished, suggesting that sensitivity in the injured hand may not have been normal before the accident (smokers,

manual workers with calloused hands). Perhaps, this resulted in poorer results of nerve repair in this study? 4.2. Comparison with results in the literature The results of this Revolnerv1 series can be compared to series of primary repair of the digital nerves by epineural suture because existing studies of nerve regeneration conduits published to date mainly concern the treatment of loss of nerve tissue. Tables 5, 7 and 8 compare our results to the series by Altissimi et al. [2], Efstathopoulos et al. [9], Elias et al. [14] and Wang et al. [15], globally and for the subgroups of jagged lacerations or sharp transections. The series by Altissimi et al. [2] involved 54 direct epineural sutures for digital nerve repair with a minimum follow-up of one year. They obtained more good and excellent results than in our study (50% vs 41%). This difference may be due to the greater number of children in their cohort, fewer jagged lacerations (17% vs 41%) as well as the longer follow-up (1 year). Our results for sensitivity after a 6-month follow-up are therefore not as good as those obtained by Altissimi et al. at one year. Efstathopoulos et al. [9] obtained good or excellent results in 48% and more poor results. If we consider that the number of children in that series and the limited number of jagged Table 7 Results of the static two-point discrimination Weber test (WS). Comparison with the Wang series.

Sharp transection Average WS Sharp transection Proportion of WS  7 mm Jagged lacerations Average WS Jagged lacerations Proportion of WS  7 mm

Revolnerv1

Wang

9 (n = 16)

6 (n = 29)

50%

62%

12 (n = 11)

8 (n = 37)

10%

43%

A. Arnaout et al. / Chirurgie de la main 33 (2014) 279–285

284

Table 8 Results of the static two-point discrimination Weber test (WS). Comparison with the Elias series.

Revolnerv1 Global WS (n = 27) Elias Global WS (n = 64) Revolnerv1 Jagged lacerations (n = 11) Elias Jagged lacerations (n = 44) Revolnerv1 Sharp transections (n = 16) Elias Sharp transections (n = 39)

Very good WS  6 mm (%)

Good WS 7–10 mm (%)

Poor WS 11–15 mm (%)

22

19

55

30

36

34

0

18

82

32

25

43

38

19

63

28

49

23

lacerations could have helped improve the results, we can conclude that our results after 6 months of follow-up were similar to theirs and will probably be better after longer follow-up. Elias et al. [14] obtained more good and very good results than in our study, except in patients over 40 (identical percentages) and for sharp transections (Revolnerv1 superior). Nevertheless, in their personal grading system, results that were classified as fair in our series (11–15 mm) were considered to be poor in their series, the population was younger (mean age 30 years old vs 38 in our Revolnerv1 series), while there were more patients and longer follow-up. Wang et al. [15] studied the results of the WS test in a series of 90 digital nerve repairs. The cohort included 67 adult patients seen for a minimum follow-up of 1 year. The results of the WS test in the different subgroups in that study seemed better (mean WS was lower and proportion of results  7 mm higher) than in ours. Nevertheless, the follow-up (6 months vs 1 year) was short and the number of patients in each subgroup was low in our Revolnerv1 series. 4.3. Influence or arterial repair on the results [16] With good or excellent results in 59%, recovery of sensitivity seemed to be best in the group with no associated palmar artery injury. Arterial repair was only performed in 3/10 of the patients with palmar digital artery injury, which probably influenced the potential for sensory recovery in this subgroup. Indeed, the importance of repair of the palmar digital artery for nerve recovery is known [16]. 4.4. Complications There were no complications due to the Revolnerv1 conduit in our study. These results confirm those in other series on collagen conduits. There were no rejections of the Neuragen1 conduit (collagen type1) in the study by Taras et al. [17] in 74 nerves, only scar sensitivity. The retrospective study by Thomsen et al. [10] of 10 digital neuromas treated by Revolnerv1 did not report any cases of poor tolerance or allergies.

5. Conclusion The goal of this study was to evaluate the advantages of using Revolnerv1 as a guide for nerve regeneration and to protect direct sutures of the palmar digital nerves in particular in jagged lacerations. After 6 months follow-up, no complications were reported from Revolnerv1, good or excellent results were obtained in 41%, recovery of sensitivity was satisfactory and our results confirm those in other series of primary digital nerve repair. However, Revolnerv1 was not shown to be significantly better for jagged lacerations, there was merely a tendency towards better sensory results. Disclosure of interest The authors declare that they have no conflicts of interest concerning this article. Acknowledgements Eight surgeons participated in this study: Ahlam Arnaout, Charlotte Brulard, Christophe Chantelot, Thomas Gahagnon, Patrick Leps, Alexandre Sauvage, Alexandre Stoven, Guillaume Wavreille. We would like to thank them for their willingness to help. References [1] Chaise F. Plaies des nerfs collatéraux palmaires des doigts. Cahiers d’enseignement de la SOFCOT no 64. Paris: Expansion Scientifique; 1997: 106–14. [2] Altissimi M, Mancini GB, Azzarà A. Results of primary repair of digital nerves. J Hand Surg Br 1991;16:546–7. [3] Millesi H. Factors affecting the outcome of peripheral nerve surgery. Microsurgery 2006;26:295–302. [4] Hudson TW, Evans GR, Schmidt CE. Engineering strategies for peripheral nerve repair. Orthop Clin North Am 2000;31:485–98. [5] Alluin O, Wittmann C, Marqueste T, Chabas JF, Garcia S, Lavaut MN, et al. Functional recovery after peripheral nerve injury and implantation of a collagen guide. Biomaterials 2009;30:363–73. [6] Cheng CJ. Synthetic nerve conduits for peripheral nerve reconstruction. J Hand Surg 2009;34:1718–21. [7] Keilhoff G, Stang F, Wolf G, Fansa H. Biocompatibility of type I/III collagen matrix for peripheral nerve reconstruction. Biomaterials 2003;24:2779–87.

A. Arnaout et al. / Chirurgie de la main 33 (2014) 279–285 [8] Khor E. Methods for the treatment of collagenous tissues for bioprostheses. Biomaterials 1997;18:95–105. [9] Efstathopoulos D, Gerostathopoulos N, Misitzis D, Bouchlis G, Anagnostou S, Daoutis NK. Clinical assessment of primary digital nerve repair. Acta Orthop Scand Suppl 1995;264:45–7. [10] Thomsen L, Bellemère P, Loubersac T, Gaisne E, Poirier P, Chaise F. Treatment by collagen conduit of painful post-traumatic neuromas of the sensitive digital nerves: a retrospective study of ten cases. Chir Main 2010;29:255–62. [11] Comtet JJ. La sensibilité : physiologie, examen, principes de la rééducation et de la sensation. Ann Chir Main 1987;6:230–8. [12] Sullivan DJ. Results of digital neurorraphy in adults. J Hand Surg Br 1985;10:41–4.

285

[13] Al-Ghazal SK, Kierman MC, Khan K, Mc Cann J. Results of clinical assessment after primary nerve repair. J Hand Surg Br 1994;19:255–7. [14] Elias BE, Tropet Y, Brientini JM, Vichard P. Résultats de la réparation primaire de nerfs digitaux palmaires. Ann Chir Main 1994;13:107–12. [15] Wang WZ, Crain GM, Baylis W, Tsai TM. Outcome of digital nerve injuries in adults. J Hand Surg Am 1996;21:138–43. [16] Piquet M, Obert L, Laveaux C, Sarlieve P, Vidal C, Tropet Y, et al. Influence of palmar digital artery patency on neurological recovery of palmar digital nerve lesions. Chir Main 2010;29:94–9. [17] Taras JS, Nanavati V, Steelman P. Nerve conduits. J Hand Ther 2005;18: 191–7.

Sensory recovery after primary repair of palmar digital nerves using a Revolnerv(®) collagen conduit: a prospective series of 27 cases.

Despite advances in microsurgery, digital nerve repair remains a challenge due to the lack of reproducible procedures with satisfactory functional res...
871KB Sizes 0 Downloads 5 Views