Data analysis

Muscle contraction data generated from suction-type and bipolar loop–type electrodes were compared with muscle twitch data obtained using the nerve clamp electrode. Peak amplitude and 10–90% decay time from twitch tension recordings were calculated for all three electrodes. To prevent direct stimulation of the hemidiaphragm muscle preparation, stimulation intensity was preset to a maximum of 7.5 V. Muscle paralysis was subsequently recorded in the presence of BoNT/A. Twitch tension peak amplitude and 10–90% decay times were compared using a two-factor ANOVA. Each peak amplitude and 10–90% decay time reported reflect mean ± SD of twitch recordings from 7–11 separate muscles.

Results and discussion

Hemidiaphragm muscle twitch tension amplitudes on the order of 0.5–6 g for mouse diaphragm, 10–40 g for guinea pig diaphragm, 20–60 g for rabbit diaphragm, and 4–18 g for rabbit external intercostal muscle were recorded following stimulation of the nerve with the described nerve clamp–stimulating electrode device (see Figure 1). Representative control twitch tension recordings are shown (Figure 3A) before BoNT/A application. The shape, twitch amplitudes, and time course of decay were very similar to results obtained using bipolar loop and suction electrodes for mouse, guinea pig, and rabbit diaphragm. These values have been summarized in Table 1. Due to the short length of motor nerve innervating rabbit external intercostal muscle, it was not possible to stimulate this indirectly using bipolar loop (data not shown) or suction electrodes (Figure 3B). Control experiments for all muscles showed stable twitch tensions during the time course of study; tensions decreased to only 70% of their initial value after ~8.5 h in the tissue bath (data not shown). None of the twitch tension amplitudes from control experiments decreased to 50% of their initial value after almost 9 h in the tissue bath. The first twitch tension recordings following application of BoNT/A to the tissue bath showed decrease in the twitch tension amplitude (Figure 4A), suggesting muscle contraction was indeed the result of indirect muscle stimulation through the nerve. Muscle preparations in the absence of BoNT/A maintained their tension throughout the duration of the recording period (data not shown). Traces shown in Figure 4A represent 10-min intervals. If the muscle was stimulated directly, BoNT/A would have no effect on the muscle contraction amplitude. Figure 4B shows the dose-dependent effect of BoNT/A (1 pM and 1 nM) to excised rabbit external intercostal muscle as a function of time. Intercostal hemidiaphragm muscle preparations in two separate experiments were stimulated indirectly through the external intercostal nerve using the described nerve clamp electrode. All test concentrations of BoNT/A resulted in complete paralysis of the muscle during the same time course of study. BoNT/A exhibited dose dependence, where the higher BoNT/A test concentration (1 nM) gave the shortest times to reach 50% inhibition, and a smaller concentration (1 pM) resulted in a longer time to 50% inhibition and subsequently complete muscle paralysis. This is in agreement with many previous findings using bipolar loop stimulating electrodes.

Figure 3. Indirect muscle twitch tension recordings stimulated using the nerve clamp electrode. (Click to enlarge)

Figure 4. Stimulation and paralysis of rabbit intercostal muscle by BoNT/A (1 nM). (Click to enlarge)

Table 1. Peak amplitude and decay time of twitch tensions recorded from the various stimulating electrodes. (Click to enlarge)

Bipolar loop electrodes are more advantageous than suction-type electrodes as they allow stimulation of larger-diameter nerve segments, typically with excess adipose tissue adhered to the axon tract. However, they are also not without their limitations. The nerve must be long enough (>15 mm) to tie a ligature around the proximal end in order to thread it through the platinum wire loops. Likewise, suction electrodes require sufficient length to apply negative pressure to maintain the nerve inside the tubing shaft of the electrode. The negative pressure created by suction electrodes can cause the nerve to pull away from the muscle, leading to denervation.

The intent of this investigation was to design and construct a unique electrode capable of eliciting muscle contractions indirectly, regardless of anatomical constraints. Typically, the common peroneal nerve-extensor digitorum longus, tibial nerve-soleus, and phrenic nerve-hemidiaphragm muscle preparations are used to indirectly stimulate muscle because they meet this nerve length requirement. This nerve clamp–stimulating electrode possesses several advantages over currently available stimulating electrodes. The described nerve clamp electrode performs similarly to commonly used electrodes, and data obtained from its use are consistent with data obtained from other electrodes. Peak amplitude of the muscle twitch tension as well as decay time did not significantly differ among all three electrodes tested. It also is not restricted by the physical characteristics of the nerve. Any nerve length or thickness can be securely and reliably contacted by the nerve clamp without directly contacting the muscle itself. It represents an improvement over commonly used electrodes, as it is able to indirectly stimulate muscles in preparations where commonly used electrodes fail, and is compatible with existing commonly used physiology laboratory stimulation and data recording hardware.