Interfaces with the Peripheral Nerve for the Control of Neuroprostheses - International Review of Neurobiology - page 72

Several alternative designs have been proposed for improving the
amount of axonal regeneration through nonobstructive, regenerative mul-
tielectrodes. A simple design was to insert 18 needle electrodes transversally
in a nerve guide used to bridge the sectioned nerve. Initial studies in vivo
showed that such electrode arrays placed in the path of regenerating nerve
fibers allowed the recording of action potentials from as early as 8 days
postimplantation to as long as 3 months in a low proportion of the animals
( Garde, Keefer, Botterman, Galvan, & Romero, 2009 ).
The natural tendency of regenerating axons to form new small fascicles
makes optimal selective recording or stimulation difficult to accomplish.
Moreover, at difference with the normal topography of the peripheral nerve,
regenerated fascicles contain a mix of sensory and motor fibers directed to
diverse targets, with loss of the normal somatotopic map. In vitro attempts
have been made to guide regenerating axons into small fascicles by promot-
ing neurite growth into microchannels with stepping bifurcations
( Wieringa, Wiertz, De Weerd, & Rutten, 2010 ), and to separate distinct
functional types of regenerating axons by exposure to different neurotrophic
factors in Y-shape channels ( Lotfi, Garde, Chouhan, Bengali, & Romero-
Ortega, 2011 ).
New alternatives in the development of regenerative electrodes have
explored the potential benefits of designing, facilitating regenerative scaf-
folds ( Clements et al., 2013 ) and the so-called microchannel electrodes
( Fig. 2.1 G; FitzGerald, Lacour, & Fawcett, 2008 ). These electrodes could
be thought as an evolution of sieve electrodes in which, instead of growing
through holes, the axons grow via thin, narrow parallel tubes with embed-
ded electrodes. In order to resemble the topographical organization of a
healthy nerve, the tubes can incorporate guidance cues to improve nerve
regeneration directing axonal growth thus mimicking the fascicle distribu-
tion of a peripheral nerve ( Lundborg & Kanje, 1996 ). The long contacts the
electrodes make with the axon along the microchannel facilitate that every
electrode will eventually contact one or more nodes of Ranvier, increasing
the amplitude of recorded action potentials ( Lacour et al., 2009 ) and all-
owing for stimulation of single motor units with lower thresholds than with
other types of electrodes ( FitzGerald et al., 2012 ). Some studies in rats
reported that these new approaches allow a high selectivity in recording
and stimulation of the regenerated axons ( Clements et al., 2013;
Delivopoulos, Chew, Minev, Fawcett, & Lacour, 2012 ), although successful
regeneration occurred only in a low proportion of the animals ( FitzGerald
et al., 2012 ).