Key points This study examines conduction in peripheral nerves and its use dependence in tetrodotoxin‐resistant (TTXr) sodium channel (Nav1.8, Nav1.9) knockout and wildtype animals. We observed use‐dependent decreases of single fibre and compound action potential amplitude in peripheral mouse C‐fibres (wildtype). This matches the previously published hypothesis that increased Na/K‐pump activity is not the underlying mechanism for use‐dependent changes of neural conduction. Knocking out TTXr sodium channels influences use‐dependent changes of conductive properties (action potential amplitude, latency, conduction safety) in the order Nav1.8 KO > Nav1.9KO > wildtype. This is most likely explained by different subsets of presumably (relatively) Nav1.7‐rich conducting fibres in knockout animals as compared to wildtypes, in combination with reduced per‐pulse sodium influx. Abstract Use dependency of peripheral nerves, especially of nociceptors, correlates with receptive properties. Slow inactivation of voltage‐gated sodium channels has been discussed to be the underlying mechanism – pointing to a receptive class‐related difference of sodium channel equipment. Using electrophysiological recordings of single unmyelinated cutaneous fibres and their compound action potential (AP), we evaluated use‐dependent changes in mouse peripheral nerves, and the contribution of the tetrodotoxin‐resistant (TTXr) sodium channels Nav1.8 and Nav1.9 to these changes. Nerve fibres were electrically stimulated using single or double pulses at 2 Hz. Use‐dependent changes of latency, AP amplitude, and duration as well as the fibres’ ability to follow the stimulus were evaluated. AP amplitudes substantially diminished in used fibres from C57BL/6 but increased in Nav1.8 knockout (KO) mice, with Nav1.9 KO in between. Activity‐induced latency slowing was in contrast the most pronounced in Nav1.8 KOs and the least in wildtype mice. The genotype was also predictive of how long fibres could follow the double pulsed stimulus with wildtype fibres blocking first and Nav1.8 KO fibres enduring the longest. In contrast, changes in spike duration were less pronounced and displayed no significant tendency. Thus, all major measures of peripheral nerve accommodation (amplitude, latency and durability) depended on genotype. All use‐dependent changes appeared in the order NaV1.8 KO > NaV1.9 KO > wildtype, which is most likely explained by the relative contribution of Nav1.7 varying in the same order and the amounts of per‐pulse sodium influx expected in the opposite order.