Concentric versus eccentric isokinetic strengthening of the rotator cuff. Objective data versus functional test.

Twenty-two male and female college varsity tennis players trained for 6 weeks, one group using eccentric isokinetic internal and external shoulder rotation, and the second group using concentric isokinetic internal and external shoulder rotation. Subjects pretested and posttested both concentrically and eccentrically, so that training overflow and specificity could be examined. Three maximally hit tennis serves made before and after training, which were analyzed by high speed cinematography to obtain ball velocity, served as a functional performance measurement. Statistical analysis of peak torque (newton meters) and peak torque to body weight ratio have revealed significant concentric strength gains (P less than 0.005) in the concentric as well as the eccentric training groups. Eccentric strength gains were demonstrated by the concentric training group at selected speeds (P less than 0.05 and P less than 0.005) but were not generated in the eccentric group at the P less than 0.05 significance level. Functional test analysis shows an increase in maximal serve velocity at a significance level of P less than 0.005 in the concentric training group, with no significant (P greater than 0.01) increases in the eccentric group.