The effect of corticosteroid on collagen expression in injured rotator cuff tendon

Subacromial corticosteroid injections are commonly used in the nonoperative management of rotator cuff disease. The effects of corticosteroid injection on injured rotator cuff tendons have not been studied. Our aims were to characterize the acute response of rotator cuff tendons to injury through the analysis of the type-III to type-I collagen expression ratio, a tendon injury marker, and to examine the effects of corticosteroid on this response.

Sixty Sprague-Dawley rats were randomly assigned to four groups: control, tendon injury, steroid treatment, and tendon injury and steroid treatment. Six rats served as sham controls. Unilateral tendon injuries were created with full-thickness defects across 50% of the total width of the infraspinatus tendon, 5 mm from its humeral insertion. Steroid treatment with a single dose of methylprednisolone (0.6 mg/kg), equivalent to that given to humans, was injected into the subacromial space under direct visualization. Steroid treatment followed the creation of an injury in the rats in the injury and steroid treatment group. At one, three, and five weeks after the injury, the total RNA isolated from tendons was quantified with real-time polymerase chain reaction with use of primers for type-I and type-III collagen and ribosomal 18s RNA.

The type-III to type-I collagen expression ratio remained at baseline at all time-points in the control and sham groups. At one week, the type-III to type-I collagen expression ratio increased more than fourfold above the control level in the tendon injury group (p = 0.017) and the tendon injury and steroid treatment group (p = 0.003). The ratio remained greater than twofold above the control at three weeks in both groups (p = 0.003 and p = 0.037) and returned to baseline at five weeks. Interestingly, the group that had steroid treatment only showed an increase of >4.5-fold (p = 0.001) in the type-III to type-I collagen expression ratio, without structural injury to the tendon. This ratio returned to baseline levels by three weeks.

A single dose of corticosteroid does not alter the acute phase response of an injured rotator cuff tendon in the rat. However, the same steroid dose in uninjured tendons initiates a short-term response equivalent to that of structural injury.