Back to 2013 Annual Meeting Program
Outcomes Evaluation Modalities: The Pursuit to Identify A “Toolbox” in Conducting Peripheral Nerve Research
Jing Rui, MD; Robert J. Spinner, MD; Michael J. Yaszemski, MD, PhD; Anthony J. Windebank, MD; Huan Wang, MD, PhD
Mayo Clinic, Rochester, MN, USA
Purpose: The current study aims to explore the correlation of variables in four classes of nerve regeneration evaluation modalities, electrophysiology, muscle histology, neuromorphometry and functional tests, and to define variables that can serve as first choice outcomes measures.
Methods: Data were collected from 25 rats that underwent conduit repair of a 1-cm gap in the sciatic nerve. Four classes, a total of 22 attributes, were applied to evaluate regeneration: video motion analysis (dynamic, every 4 weeks; 6 variables), nerve conduction study (dynamic, every 4 weeks; 6 variables), muscle histology (endpoint; 4 variables) and neuromorphometry (endpoint; 6 variables). These attributes reflect overall sciatic function (4 variables), tibial function (9 variables) and peroneal function (9 variables), respectively. Spearman rank order correlation coefficients were used to test the association between the 22 variables. In order to seek attributes that can potentially reflect the collective function, each variable was given a rank among the 25 animals based on the values that were from best to worst. Correlations between the rank of each attribute and the sum of the ranks of all the attributes were examined using Pearson’s correlation coefficients.
Results: With alpha set to 0.05, the sample size of 25 provided the correlation analysis a 95% statistical power to detect a correlation coefficient of 0.396 (in absolute value). Significant correlations between variables within a class and of different classes, and between peroneal and tibial function attributes were revealed (0.396 < r ≤ 0.5: 17 pairs; 0.5 < r ≤ 0.7: 19 pairs; 0.7 < r < 1: 4 pairs) (table 1). Three pairs including 6 attributes had the most number of significant correlation with other attributes (peroneal nerve total myelinated fiber number/PTMF: 9; tibial nerve total myelinated fiber number/TTMF: 7; peroneal CMAP amplitude/PAMP: 7; tibial CMAP amplitude/TAMP: 6; anterior tibial muscle weight/AWR: 5; gastrocnemuis muscle weight/GWR: 5). Rank sum correlation coefficient revealed 12 attributes of significant correlation with the whole function, the top 3 being PAMP, PTMF and AWR (r > 0.7). Motion analysis attributes had the least number of correlations with other modalities.
Conclusions: Both intra-class and inter-class correlations existed. CMAP amplitude, muscle weight and total myelinated nerve fiber count can be the representative attributes for electrophysiology, histology and neuromorphometry, respectively. Motion analysis is a separate but important evaluation modality to assess the different aspect of nerve regeneration.
Back to 2013 Annual Meeting Program