Although reverse total shoulder arthroplasty (TSA) may restore shoulder abduction and forward flexion in the setting of a massive rotator cuff tear, the ability to use the extremity for ADL’s is often limited by external rotation weakness. Even though the reverse TSA restores abduction, the patient may be unable to bring the hand to his or her mouth because with the elbow flexed the weight of the hand causes the shoulder to fall into internal rotation. Concomitant transfer of latissimus dorsi (LDT) to the posterior greater tuberosity is a solution advocated by some surgeons. It is hypothesized that this inferiorly-directed force partially counteracts the superiorly-directed force of the deltoid, resulting in decreased shear forces on the glenoid baseplate-bone interface.

Three cadaver shoulder specimens were dissected and implanted with the reverse TSA. The rotator cuff was completely released to simulate a massive rotator cuff tear. Each shoulder was mounted in a shoulder controller that simulates neuromuscular control and replicates in vivo glenohumeral kinematics. The controller utilizes an optical, three dimensional tracking system. The humerus was weighted to simulate the full mass of the upper extremity and stepper motors were connected to the insertion points of the anterior, middle and posterior divisions of the deltoid by Spectra^® cord. Simulated active abduction in the scapular plane was performed using position closed-loop feedback control. The joint reaction force at the glenosphere was measured at 5° intervals from 30°–70°. A fourth stepper motor was then connected to the greater tuberosity with 2.73kg applied to simulate a LDT and the test was repeated. Five trials were performed under each condition. Four-factor ANOVA statistical analysis with Bonferroni correction and α = 0.05 was performed.

After simulated LDT the JRF demonstrated an increase in magnitude at abduction angles between 30° and 65° inclusive (p=0.033). The superiorly-directed shear force was significantly decreased as a result of the LDT between 45° and 70° (p< 0.0001). The compressive component of the JRF was increased for all abduction angles (p=0.025). The force required to achieve abduction increased for the middle deltoid (p=0.035) and anterior deltoid (p=0.036) for the simulated LDT condition at all abduction angles. The posterior deltoid force required for abduction decreased at all abduction angles (p=0.031).

In this model of reverse total shoulder arthroplasty concomitant transfer of latissimus dorsi decreased the superiorly-directed shear force. In addition to providing improved external rotation strength, these lower shear forces may have a protective effect on baseplate fixation by reducing the risk of failure in shear. This may provide additional justification for the transfer. Although superior shear was decreased, total JRF was increased as a result of an increase in the compressive component. Further investigation is needed to determine the potential gain in joint stability and whether the glenoid bone can support such elevated compressive forces. Additionally, the force required in the anterior and middle deltoid was increased after the LDT. This indicates the need for sufficient deltoid strength and rehabilitation.