Test-retest reliability of the soleus H-reflex is affected by joint positions and force levels

The H-reflex has been extensively used to exam the spinal motoneuron excitability. However, no report has demonstrated the test-retest reliability of the H-reflex assessment during voluntary muscle contractions and its relation to joint positions. The purpose of this study was to determine the test-retest reliability of the soleus (SOL) H-reflex during rest and 10, 30, 50% maximal voluntary contractions (MVC) at the ankle joint angles of neutral (0deg), plantarflexion (20deg), and dorsiflexion (-20deg), respectively in a sitting position. Ten healthy participants (age 24.9 ± 5 years, height 168.3±8.8 cm, weight 62.7±12.3 kg) were tested for the SOL H-reflex on two separate occasions within 7 days. The SOL H-reflex was elicited by electrical stimulation applied to the posterior tibial nerve. The maximal amplitude of H-reflex wave (Hmax) and M-wave (Mmax) were recorded during rest and three submaximal voluntary contraction levels and at three ankle joint positions in a randomised order. The ratios of Hmvc /Hrest and Hmax/Mmax were also calculated. The Intraclass Correlation Coefficient (ICC2,1, SPSS) was used to examine the test-retest reliability. Test-retest reliability of the SOL H-reflex during rest was excellent at 0deg (ICC = 0.92) and 20deg plantar flexion (ICC = 0.96) and was good at -20deg dorsiflexion (ICC = 0.75). Inconsistent ICC values were found during submaximal voluntary contractions at three ankle joint positions (range from 0.62 to 0.97), with the highest values observed during 10%MVC (0deg: 0.92, 20deg: 0.93, and -20deg: 0.95). Moderate ICCs were found in Hmax/Mmax (0deg: 0.80, 20deg: 0.92, -20deg: 0.79). A much greater variation in ICC was observed in other comparisons (from -1.66 to 0.977). In conclusion, repeated measurements of SOL H-reflex demonstrated a high reliability at the neutral and plantar flexion positions than that at the dorsiflexion position during rest. The results also indicated that the SOL H-reflex during 10% MVC was more reliable than that at 30 and 50 %MVC. The reliability of the SOL H-reflex appeared to be significantly affected by levels of voluntary contractions and ankle positions.