212 B. Hiebl, S. Bog, R. Mikut, C. Bauer, O. Gemeinhardt, F. Jung, T. Krüger Applied Cardiopulmonary Pathophysiology 14: 212-219, 2010 In vivo assessment of tissue compatibility and functionality of a polyimide cuff electrode for recording afferent peripheral nerve signals B. Hiebl1, S. Bog2, R. Mikut3, C. Bauer3, O. Gemeinhardt4, F. Jung1, T. Krüger5 1 Center for Biomaterial Development and Berlin-Brandenburg Center for Regenerative Therapies, Helmholtz-Zentrum Geesthacht, Centre for Materials and Costal Research, Teltow, Germany; 2University Heidelberg, Interdisciplinary Biomedical Research Facility, Heidelberg, Germany; 3Karlsruhe Institute of Technology (KIT), Institute for Applied Computer Science, Karlsruhe, Germany; 4Institute of Veterinary Anatomy, Faculty of Veterinary Medicine, Freie Universität Berlin, Germany; 5inomed Medizintechnik GmbH, Teningen, Germany Abstract Minimally-invasively implantable spiral cuff electrodes coated with polyimide (PI) for insulation in combination with suitably designed amplifiers were recently reported to allow signalling of electroneurograms (ENGs). These cuff electrodes can be used to record the respiratory drive after implantation on the phrenic nerve, but PI can cause epineurial fibrosis, fiber loss, and limited reproducibility of recordings. This study aimed to explore the tissue reaction in response to a flexible tripolar cuff electrode embedded in a thin (10 µm) PI insulating carrier after implantation around the sciatic nerves of rats (n=4) at the branching into the N. peroneus communis, the N. tibialis and N. cutaneus surae caudalis. 28 days after implantation the electrode functionality was proven and ENG signals recorded. In addition, transverse sections of the implantation site were analysed after hematoxylin-eosin (HE) and Weigert´s fibrin staining for changes in the tissue morphology and the number of myelinated nerve fibres. 28 days after implantation the electrode contact to the nerve still was sufficient for signal recording and no changes in the nerve morphology and the number of myelinated nerve fibres could be noted. However, cuff functionality for nerve signal transfer was limited by a fibrous capsule, which covered the whole electrode and which was composed mainly of tightly packed fibroblasts and fibrin. The formation of such a fibrous capsule is known to be caused by the foreign body response and shows a limited tissue compatibility which might be mainly related to the PI insulating material. Further studies will address the investigation of alternative elastic matrix-materials to achieve a strong integration of the electrode in the nerve tissue and in this way a long term functionality of the cuff electrode. Key words: polyimide, cuff sensor, peripheral nerve, neural damage, functional electric stimulation In vivo assessment of tissue compatibility and functionality of a polyimide cuff electrode ... Introduction The respiratory drive of anesthetized animals can be measured by using electroneurograms (ENGs) which are obtained from the phrenic nerve using minimally-invasively implantable spiral cuff electrodes [1]. They are advantageous compared to invasive nerve electrodes (e.g. hook electrode), because they can provide signals without cutting or desheathing of the nerve [2-10]. The spiral cuff electrode is designed to wrap around the nerve and, due to its self-sizing property, it adapts its diameter to the size of the nerve [4]. However, ENG recordings from cuff electrodes have a low signal-to-noise ratio and are often affected by artefact signals, mostly generated by the muscles nearby [11]. Artefact reduction might be achieved by using proper amplifiers [11-12]. The long-term efficacy of the cuff electrode for ENG measurements depends on the tissue-material interaction especially the chronic inflammatory response induced by the electrode material. Polyimide (PI) is broadly explored as membrane or particle forming biomaterial [13-15] as well as substrate and insulation material for neural implants due to its chemical and mechanical durability [1619], but epineurial fibrosis, fiber loss, limited reproducibility of recordings and variability of stimulation conditions have been documented after extraneural cuff electrode implantation [20-21]. In this study we explored the foreign body response/tissue compatibility of adult female rats in response to a new flexible tripolar cuff electrode design composed of a thin (10 µm) and flexible PI insulating carrier and three circumneural platinum electrodes after 28 days of implantation at the branching of the sciatic nerve into the N. peroneus communis, the N. tibialis and N. cutaneus surae caudalis. 213 Materials and methods Polyimide cuff electrode Nerve signals were recorded from the sciatic nerve with a tripolar cuff electrode which consisted of a cuff electrode and an electrode extension cable (length 150 mm, fig. 3). The electrode to electrode distance in the cuff was 1.5 mm. The fabrication of the polyimide-based microstructure with a single metallization layer and a PI coating (Pyralin PI 2611, HD Microsystems, Bad Homburg, Germany) has been described in detail previously [12, 18]. Animals The study was approved by the regional council of Karlsruhe (Germany) and performed in accordance with the EN DIN ISO 10993-6. A total of four female rats of the inbred strain BDX were used in the present study. The animals were provided from a commercial breeder (Charles River, Germany). They were 7 ± 1 months old and had a body weight of 220 ± 10 g before surgery. Housing and animal care The animals were group-housed in MacrolonTM-based type III cages with at least two animals per cage and ad libitum supply of water and food to guarantee homeostasis as well as exclude acute and chronic medical conditions. According to the European Guidelines for care and use of laboratory animals, the animals were kept at a temperature of 20 ± 1 °C and a humidity of 55%. Artificial lighting was used to maintain a rhythm of 12 h day and 12 h night. Surgical procedure For electrode implantation, animals were anaesthetized with isoflurane (1.5 v/v%). After shaving and disinfecting the dorsal neck 214 B. Hiebl, S. Bog, R. Mikut, C. Bauer, O. Gemeinhardt, F. Jung, T. Krüger region and the right lateral Regio femoris with poly(1-vinyl-2-pyrrolidon)iodine complex (Betaisodona®, Mundipharma, Germany), firstly, the sciatic nerve was freed from surrounding tissue via an incision of about 20±5 mm through the skin and the Musculus biceps femoris. The initially tube shaped cuff was bent with forceps and positioned behind the nerve. Thereafter the forceps were removed allowing the cuff to wrap around the branching of the sciatic nerve into the N. peroneus communis, the N. tibialis and N. cutaneus surae caudalis avoiding compression and stretching. After release, the spiral cuff was closed covering the whole nerve perimeter, to which it adhered by surface tension. The electrode interconnect line was passed through the incision of the Musculus biceps femoris avoiding tension. The connector A bundle (see fig. 2) was placed and fixed by a single stitch subcutaneously on the upper lateral side of the hind limb. The incision of the muscle was closed by a continuous suture using blue dyed poly(lactide-co-glycolide) (VicrylPlus, 3-0, Ethicon, Germany) and the sensor bundle with the connector B (see fig. 2) was subcutaneously placed on the back and exit out of the body through an incision at the prepared neck region which was subcutaneously stabilised by a polypropylene mesh (15×15mm, Parietene®, Sofradim, Germany). To close the skin at the surgical site of the limb and the neck, suture clips (Michel Suture Clip, 7.5 mm, Harvard Apparatus, USA) were used. Additionally a rat jacket (Harvard Apparatus, USA) was used to protect the surgical site at the neck and also to protect the animal in the post surgical or treatment situation. Electrode functionality testing The functionality of the electrode to record afferent ENG signal of the rat sciatic nerve was tested after mechanical stimulation of the rat’s paw using a 6.10 Semmes-Weinstein filament representing a force of linear pressure corresponding to a mass of 75 g. [22]. The method to investigate the functionality of the cuff electrode especially recording of nerve signals has been described in detail previously [12]. Histology Four weeks (28 days) after the implantation, the rats were sacrificed by the use of CO2. The skin of the implantation site was opened in the direction of the former incision, blood was swabbed, and the electrode cuff together with the surrounding tissue (see fig. 3) were immediately placed in a small jar containing the fixing agent, a buffered neutral 4 v/v% formalin solution (pH 7) for further histological examinations. Histological semi-sections were made for light microscopic analysis (phase contrast mode) after acrylic resin embedding (Technovit 7200 VLC, Kulzer, Germany). Sections were prepared in 5 µm thickness and subjected to staining with hematoxylin-eosin (HE) and Weigert´s fibrin according to the protocol of Romeis [23]. Five slides per animal were prepared and each slide was evaluated at five different fields of view with a transmitted light microscope in the phase contrast mode (Zeiss, Germany) using the image analysis software AxioVisio (Zeiss, Germany). Statistics Results are expressed as mean value ± standard deviation for continuous variables of at least three independent experiments. Differences were assessed using two-tailed Student´s t-tests for unpaired samples. In case of multi-sample comparison, variance analysis was performed. Significance was assumed if p value was less than 0.05. In vivo assessment of tissue compatibility and functionality of a polyimide cuff electrode ... Results Electrode functionality testing The implantation of the cuff electrode around the sciatic nerve took about 1-2 min. Immediately after the implantation, the contact interface between the electrode and the nerve was sufficient to record afferent ENG signals (sampling frequency 50 kHz) after stimulation with defined mechanical stimuli on the naked skin of the paw. With a 6.10 SemmesWeinstein filament, a pressure was applied approximately once per second inducing afferent signals in response to the mechanical stimulus (see figure 1). Since it was difficult to 215 detect the raw signal, a filtered activity signal was generated by computing absolute values of the raw signal, followed by an infinite-input response filter of first order (filter constant a = 0.999). The induced afferent nerve signals were seen in the filtered activity signal. At approximately 10.5 s, an external disturbance occurred. In addition, a spectrogram (windows size 1024 sample points) of the raw signal showed the signal power for the different frequencies (fig. 2). Afferent nerve signals ranged up to 7 kHz. All analyses were performed with the open-source Matlab toolbox Gait-CAD [24-25]. Figure 1: Afferent nerve ENG signal of the rat sciatic nerve after mechanical stimulation of the paw using a 6.10 Semmes-Weinstein filament representing a force of a mass of 75 g, (A): raw signal amplitude, (B): filtered afferent activity signal; at 10.5 sec a disturbance occurred. Figure 2: Spectrogram of an ENG raw signal of the rat sciatic nerve after mechanical stimulation of the paw using a Semmes-Weinstein filament in the size of 6.10 representing a force of a mass of 75 g. 216 B. Hiebl, S. Bog, R. Mikut, C. Bauer, O. Gemeinhardt, F. Jung, T. Krüger At the end of the implantation period, the electrode was still contacted to the nerve and continued to measure current impedance. However, signal recording was limited. After 28 days of implantation, the signal intensity was decreased and the impedance was increased. Macroscopical findings In all animals, 28 days after the implantation, the cuff electrode stayed at the position targeted by implantation and surrounded the sciatic nerve perimeter. No pathological finding was observed in the periimplantary subcutaneous and muscular tissues. As shown in figure 3, the cuff electrode and the electrode expansion cable including the connectors appeared covered by thin fibrous tissue, well vascularised and in continuity with the epineurium (cuff) and perimysium (connectors and expansion cable). Histological analysis Semi-thin transverse sections of the cuffed nerves (light microscopy-phase contrast mode), showed that the general morphology of the cuffed nerves was similar to that of the intact contralateral nerves (fig. 4 A and D). The transverse-sectional area of the cuffed and non-cuffed contralateral sciatic nerve at the branching into the N. peroneus communis, the N. tibialis and N. cutaneus surae caudalis clearly showed the distinct fascicles (tibial, peroneal, sural branches), and the endoneurium compartments encircled by the perineurium and the outer loose epineurium. This newly formed tissue layer covering the cuff electrode was mainly composed of fibrin (fig. 4 C), flattened fibroblasts and the newly formed layer was significantly thicker (26.7±6.8 µm, n=4) than the epineurium (fig. 5, 19.2+5.5 µm). The newly formed layer covered the whole electrode cuff surface including the apical tip of the electrode (fig. 3). The density of myelinated fibres in the peripheral and central nerve area within the cuff electrode ranged between 259 and 431 per Figure 3: Implantation site 28 days after placing of a cuff electrode on the sciatic nerve of rats (n=4). A: Spiral nerve cuff electrode wrapped around the sciatic nerve and covered by an artificial tissue layer. B: Electrode extension cable and connector (connector A) to the cuff electrode, both covered by a newly formed tissue layer. In vivo assessment of tissue compatibility and functionality of a polyimide cuff electrode ... 217 Figure 4: Transverse sectional area of the rat sciatic nerve 28 days after placing of a spiral cuff electrode on it (A) and without a cuff electrode (D, control), haematoxylin-eosin staining; (1a-c) distinct sciatic nerve fascicles: (1a) tibial, (1b) common peroneal, (1c) sural; (2) Musculus biceps femoris, (3) newly formed tissue layer covering the cuff electrode, (4) electrode cuff, (5) perimysium; (B, C) electrode-tissue interface after heamatoxylin-eosin staining (B) and fibrin staining (C); samples embedded acrylic resin, phase contrast microscopy in transmission, primary magnification 10× (A, D) and 40× (B, C). Figure 5: Sciatic nerves of rats 28 days after implantation of spiral nerve cuff electrodes (n=4); (A) Density of myelinated fibres; (B) Thickness of the epineurium and the artificial layer covering the cuff electrode, gray: nerve with cuff electrode, shaded: nerve without cuff electrode; 5 fields of view analysed per section (n=4), mean ± standard deviation. 10×103 µm2 (peripheral 331 ± 72 per 10×103 µm2, central 332 ± 72 per 10×103 µm2, n=4) and was similar to those found in the intact contralateral nerves (peripheral 350 ± 51 per 10×103 µm2, central 373 ± 58 per 10×103 µm2) (fig. 5). Discussion The study showed that a spiral cuff electrode which was placed on the sciatic nerve was able to record signals for 28 days despite the fact that the cuff electrode was covered by a thin, tightly packed layer of fibroblasts and fibrin as previously reported [33-34]. This tissue layer formation was the result of a foreign body response which is known as a widely in- vestigated chronic inflammatory response to an implant [26-28] and is important in the prevention of implant failure because the capsule minimizes adverse effects of the implant on the periimplant tissue [29]. The thickness and cellular morphology of the encapsulation tissue dependent on the shape of the implant [26-27], the surface texture of an implant [28], and the materials from which it is fabricated. Changes in the functionality of implanted electrodes to sense signals have been attributed to the growth of the fibrous tissue capsule. Formation of the encapsulation tissue changed the signal-to-noise ratio of recorded nerve compound action potentials [36], and decreased the selectivity and increased the length dependence of epimysial electrodes [37]. 218 B. Hiebl, S. Bog, R. Mikut, C. Bauer, O. Gemeinhardt, F. Jung, T. Krüger Further studies will focus on alternative materials for PI to reduce the limitations in electrode functionality. Cuff materials, which allow strong tissue integration [38] might be helpful to reduce tissue layer formation and the use of multifunctional polymers with shape-memory capabilities for cuff design [39-40] might allow shorter implantation times. Conclusion The minimally-invasively implantable spiral cuff electrode coated with polyimide (PI) for insulation could be shown in vivo to allow signal recording despite of a fibrous capsule formation between the electrodes and the nerve. For this reason the cuff electrode can be used to record electroneurograms (ENGs) for instance on the phrenic nerve to record the respiratory drive. Acknowledgments The authors thank Prof. Dr. Thomas Stieglitz, Laboratory for Biomedical Microtechnology, Department of Microsystems EngineeringIMTEK, University of Freiburg, Freiburg, Germany, for supporting the study. References 1. Sahin M et al. Spiral nerve cuff electrode for recordings of respiratory output. J Appl Physiol 1997; 83 (1): 317-22 2. Haugland KH, Hoffer JA. Slip information provided by nerve cuff signals: Application in closed-loop control of functional electrical stimulation. IEEE Trans Rehabil Eng 1994; 2: 29-36 3. Haugland HK, Sinkjaer T. Cutaneous whole nerve recordings used for correction of footdrop in hemiplegic man. IEEE Trans Rehabil Eng 1995; 3: 307-17 4. Naples GG et al. A spiral nerve cuff electrode for peripheral nerve stimulation. IEEE Trans Biomed Eng 1988; 35 (11): 905-16 5. Nikolic ZM et al. Instrumentation for ENG and EMG recordings in FES systems. IEEE Trans Biomed Eng 1994; 41 (7): 703-6 6. Popovic DB et al. Sensory nerve recording for closed-loop control to restore motor functions. IEEE Trans Biomed Eng 1993; 40 (10): 1024-31 7. Sahin M, Durand DM, Haxhiu MA. Whole nerve recordings with the spiral nerve cuff electrode. in Proceedings of the Sixteenth Annual International Conference of the IEEE Engineering in Medicine and Biology Society 1994. Baltimore IEEE 8. Stein RB et al. Principles underlying new methods for chronic neural recording. Can J Neurol Sci 1975; 2 (3): 235-44 9. Stein RB et al. Stable long-term recordings from cat peripheral nerves. Brain Res 1977; 128 (1): 21-38 10. Stein RB, Pearson KG. Predicted amplitude and form of action potentials recorded from unmyelinated nerve fibres. J Theor Biol 1971; 32 (3): 539-58 11. Demosthenous A, Triantis IF. An adaptive ENG amplifier for tripolar cuff electrodes. IEEE Journal of Solid-state Circuits 2005; 40 (2): 412-21 12. Stieglitz T, Klausmann D, Krueger TB. Development of a simple low noise amplifier for recording of sensory mass signals from peripheral nerves. Biomed Tech (Berl) 2009; 54 (1): 1-7 13. Albrecht W et al. Preparation of highly asymmetric hollow fiber membranes from poly(ether imide) by a modified dry-wet phase inversion technique using a triple spinneret. Journal of Membrane Science 2005; 262 (1-2): 69-80 14. Altankov G et al. On the tissue compatibility of poly(ether imide) membranes: an in vitro study on their interaction with human dermal fibroblasts and keratinocytes. Journal of Biomaterials Science-Polymer Edition 2005; 16 (1): 23-42 15. Trimpert C et al. Poly(ether imide) membranes modified with poly(ethylene imine) as potential carriers for epidermal substitutes. Macromolecular Bioscience 2006; 6 (4): 274-284 16. Shamma-Donoghue SA et al. Thin-film multielectrode arrays for a cochlear prosthesis. IEEE T Electron Dev 1982; 29 (1): 136-44 17. Boppart SA, Wheeler BC, Wallace CS. A flexible perforated microelectrode array for In vivo assessment of tissue compatibility and functionality of a polyimide cuff electrode ... 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. extended neural recordings. IEEE Trans Biomed Eng 1992; 39 (1): 37-42 Stieglitz T et al. Micromachined, polyimidebased devices for flexible neural interfaces. Biomed Microdevices 2000; 2 (4): 283-94 Rubehn B et al. A MEMS-based flexible multichannel ECoG-electrode array. J Neural Eng 2009; 6 (3): 036003 Grill WM, Veraart C, Mortimer JT. Selective activation of peripheral nerve fascicles: use of field steering currents. in Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 1991 Grill WM, Mortimer JT. Electrical impedance of electrode encapsulation tissue. in Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 1992 Kumar S et al. Semmes-Weinstein monofilaments: a simple, effective and inexpensive screening device for identifying diabetic patients at risk of foot ulceration. Diabetes Res Clin Pract 1991; 13 (1-2): 63-7 Mulisch M, Welsch U (Eds.). Mikroskopische Technik. 18 ed. 2010, Spectrum: Baltimore Mikut R et al. The open source Matlab toolbox Gait-CAD and its application to bioelectric signal processing. in DGBMT-Workshop Biosignal Analysis. 2008. Potsdam http://sourceforge.net/projects/gait-cad/ Wood NK, Kaminski EJ, Oglesby Rj. The significance of implant shape in experimental testing of biological materials: disc vs. rod. J Biomed Mater Res 1970; 4 (1): 1-12 Matlaga BF, Yasenchak LP, Salthouse TN. Tissue response to implanted polymers: the significance of sample shape. J Biomed Mater Res 1976; 10 (3): 391-7 Taylor SR, Gibbons DF. Effect of surface texture of the soft tissue response to polymer implants. J Biomed Mater Res 1983; 17: 205-27 Lago N et al. Assessment of biocompatibility of chronically implanted polyimide and platinum intrafascicular electrodes. IEEE Trans Biomed Eng 2007; 54 (2): 281-90 Klinge PM et al. Immunohistochemical characterization of axonal sprouting and reactive tissue changes after long-term implantation of a polyimide sieve electrode to the transected adult rat sciatic nerve. Biomaterials 2001; 22 (17): 2333-43 View publication stats 219 31. Richardson RR Jr., Miller JA, Reichert WM. Polyimides as biomaterials: preliminary biocompatibility testing. Biomaterials 1993; 14 (8): 627-35 32. Sun Y et al. Assessment of the biocompatibility of photosensitive polyimide for implantable medical device use. J Biomed Mater Res-A 2008; 90 (3): 648-55 33. McNamara A, Williams DF. The response to the intramuscular implantation of pure metals. Biomaterials 1981; 2 (1): 33-40 34. Stensaas SS, Stensaas LJ. Histopathological evaluation of materials implanted in the cerebral cortex. Acta Neuropathol 1978; 41 (2): 145-55 35. Anderson J. Mechanisms of inflammation and infection wth implanted devices. Cardiovasc Pathol 1993; 2 (3): 33S-41S 36. Lefurge T et al. Chronically implanted intrafascicular recording electrodes. Ann Biomed Eng 1991; 19 (2): 197-207 37. Grandjean PA, Mortimer JT. Recruitment properties of monopolar and bipolar epimysial electrodes. Ann Biomed Eng 1986; 14 (1): 53-66 38. Hiebl B et al. Tissue Integration of Two Different Shape-Memory Polymers with Poly(εcaprolactone) Switching Segment in Rats. Materials Research Society Symposium Proceedings 2009; 1140: 143-149 39. Lendlein A et al. Shape-memory polymers as a technology platform for biomedical applications. Expert Rev Med Devices 2010; 7 (3): 357-79 40. Behl M, Razzaq MY, Lendlein A. Multifunctional Shape-memory Polymers. Advanced Materials 2010; 22 (31): 3388-3410 Correspondence address: Bernhard Hiebl, M.D. Center for Biomaterial Development and Berlin Brandenburg Center for Regenerative Therapies Helmholtz-Zentrum Geesthacht Zentrum für Material- und Küstenforschung GmbH Kantstr. 55 14513 Teltow Germany bernhard.hiebl@hzg.de