Cervical afferents are necessary for closed‐loop epidural stimulation‐induced respiratory neuroplasticity

Abstract

Epidural electrical stimulation restores vital functions in neurological disorders such as stroke and spinal cord injury. After spinal injury, epidural stimulation improves locomotion, cardiovascular and bladder function, and trunk stability via neuromodulation of spinal neural networks. We have shown in rats that four days of epidural stimulation given via a closed‐loop paradigm (CLES) elicits spinal respiratory plasticity (Malone et al., FASEB J 2020;2021). Precise mechanisms of these effects in other sensorimotor systems are unclear, but modeling (Capogrosso et al., 2013) and rodent (Lavrov et a., 2008) studies suggest that inputs from segmental sensory afferent neurons are essential. We hypothesized that intact sensory afferents from the diaphragm (e.g. C3‐C5) are necessary for CLES‐induced respiratory plasticity. To test this, adult, female, Sprague‐Dawley rats were implanted with CLES electrodes at C4 and bilateral diaphragm recording electrodes (n=14). A subset of rats received C3‐C5 cervical dorsal rhizotomy (CDR) at the time of electrode implantation (n=6). After 1 week of recovery, rats received CLES (sub‐motor threshold epidural stimulation triggered by diaphragm EMG) ~20hr per day for 14 days. Inspiratory‐triggered spinal motor evoked potentials were recorded every 2 days. CLES for 2 weeks facilitated left diaphragm peak to peak amplitude of the stimulus‐triggered average in sham rats on all days measured (vs. day 0; all p< 0.05). Right diaphragm in sham animals only showed facilitation on day 12 vs day 0 (0.005 vs 0.003; p<0.05). In contrast, CDR rats exhibited blunted peak to peak magnitudes vs sham on all days except day 6 in the left diaphragm where there was also facilitation (vs day 0; p<0.05). Days 2, 4, 12 and 14 of peak to peak magnitude in the right diaphragm of CDR rats was significantly lower than sham (all p<0.05) and vs day 0 (all p<0.05). In addition, motor threshold (e.g. the lowest current that can elicit an evoked potential) was significantly higher in CDR vs sham rats on days 4 (14.27% ± 15.02%; p=0.005), 6 (19.23% ± 21.89% vs ‐29.43% ± 3.77%; p=0.003), and 8 (18.10% ± 16.05% vs ‐20.25% ± 10.34%; p=0.004). These results suggest that CDR may attenuate the magnitude of the evoked potential and capacity to facilitate over the course of 14 days. However, it appears some compensatory plasticity may be occurring after CDR since 1) the increase in motor thresholds return to what is seen in sham animals after day 10 of CLES and 2) facilitation of the stimulus triggered average was seen on day 6 in the left diaphragm. It is also interesting to note the differential effects of CLES on right vs. left diaphragm, highlighting the differences in innervation or possible laterality in the capacity for spinal plasticity in this setting. Ongoing immunohistochemical analyses of C3‐C5 spinal cord and diaphragm may lend more insight into the mechanistic underpinnings of these results. This work represents the first report that long‐term (i.e. 2 weeks) CLES can elicit spinal respiratory neuroplasticity in healthy, awake, freely‐behaving rats and that intact sensory afferents may be necessary for this to occur. Future studies will be essential to elucidate the therapeutic potential of neuromodulation for improvement of respiratory deficits in individuals with neurological disorders.