To the Editors:

We read with interest the letter by Beri and Khattri (1) and the article by Marie et al (2), published in recent issues of Arthritis Care & Research. Although Marie et al (2) had previously reported that statin-associated tendon toxicity might be due to 1) reduced cholesterol content of tendon cell membranes, making them unstable, or 2) reduced levels of regulatory proteins involved in the maintenance of tendon cells, Beri and Khattri (1) described that hydroxymethylglutaryl-coenzyme A blockers acting as matrix metalloproteinase (MMP) inhibitors are quite attractive, because it was reported that simvastatin inhibited MMP-9 secretion from human saphenous vein smooth muscle cells by inhibiting the RhoA/ROCK pathway (3). The possible role of MMPs in the statin-associated tendinopathy was first postulated by Pullatt et al (4), who speculated that the inhibition of MMP-9 and augmentation of tissue inhibitor of metalloproteinases 1 (TIMP-1) in macrophages by statins could impair tendon remodeling and contribute to tendinopathy.

First, the systemic effect of statins on serum MMP levels has been reported (5), but the changes of MMPs within tendons after statin use have not been previously reported in any study, including the study by Pullatt et al (4).

Second, the results of MMP expressions in tendon injuries were very conflicting in various studies (3, 6, 7). Although Pullatt et al (4) speculated that decreased MMP-9 might be related to statin-associated tendinopathy and Corps et al (6) showed that ciprofloxacin enhanced the stimulation of MMP-3 expression by interleukin-1β in human tendon-derived cells, Jones et al (7) reported that messenger RNA expressions of MMP-3 were decreased and those of MMP-9 were increased in ruptured Achilles tendon samples compared with normal tendons. Therefore, the role of MMPs on tendon matrices is still inconclusive.

Third, although Beri and Khattri (1) reported that the suppressive effect of statins on MMP-9 might be important in understanding the pathogenesis of statin-associated tendinopathy, aprotinin (a broad spectrum proteinase inhibitor including MMP inhibition) has paradoxically been used for treating patellar and Achilles tendinopathies (8).

Fourth, the eicosanoids such as prostaglandin E2 (PGE2) and leukotriene B4 (LTB4) have been implicated in the development of tendinopathy (9). Li et al showed that cyclic stretching of human tendon fibroblasts increased the production of PGE2 and LTB4, and blocking PGE2 production also led to increased LTB4 levels and vice versa (9). Because it was shown that atorvastatin suppressed PGE2 levels (10), possibly increased LTB4 levels might be implicated in the pathogenesis of statin-associated tendinopathy.

Therefore, further studies should be performed to elucidate the role of MMPs and eicosanoids in the pathogenesis of statin-associated tendinopathy and to evaluate their serial changes according to the duration of the statin use.

Jae Il Shin MD*, Jae Seung Lee MD*, * Yonsei University College of Medicine and Severance Children's Hospital, Seoul, Korea.