Each
Based in part on clinical observations of persistent stiffness, pain, and cartilage damage after prolonged immobilization, in a 1960 JBJS paper, Robert B. Salter described degenerative changes in cartilage of rabbit knee joints that had been immobilized. He suggested that this “obliterative degeneration” might be related to adherence of synovium to the articular surface, and he wondered elsewhere in the orthopaedic literature, “If intermittent motion is good for articular cartilage, would continuous motion be even better?”
This background led to the classic December 1980 JBJS publication in which Salter and his colleagues hypothesized that “continuous passive motion of a synovial joint in vivo would have a beneficial biological effect on the healing of full-thickness defects in articular cartilage.”
To test the hypothesis, Salter et al. made full-thickness cartilage defects at four sites in the knees of 147 rabbits. The rabbits were subjected postoperatively to either immobilization, intermittent active motion (normal cage activity), or continuous passive motion (CPM) created by a custom-made apparatus. Outcome measures included clinical observation of the animals, joint stiffness, and histology.
The extent of ultimate postoperative stiffness, adhesions, and cartilage healing all varied with the degree of immobilization, leading the authors to conclude that CPM
- Was well tolerated by the animals without causing harm detectable by gross or histologic evaluation
- Was associated with fewer adhesions than immobilization, and
- Stimulated more rapid and complete cartilage restoration than either immobilization or intermittent active motion.
Subsequent work by Salter and co-workers evaluated the effect of CPM on other animal models of full-thickness cartilage defects, intra-articular fractures, acute septic arthritis, patellar tendon injury, ligament repair, autogenous and allogenic periosteal and osteoperiosteal grafts, and other conditions. Based in part on the favorable results of these pre-clinical studies as well as preliminary clinical trials, Salter suggested in CORR in1989 that CPM might be indicated after a host of other orthopaedic procedures, including open reduction and internal fixation of intra-articular or selected diaphyseal and metaphyseal fractures, capsulotomy and arthrolysis for post-traumatic arthritis, synovectomy for rheumatoid arthritis or hemophilic arthropathy, arthrotomy and drainage of septic arthritis, release of contractures or adhesions, metaphyseal osteotomy with internal fixation, and reconstruction of a medial collateral ligament.
A Google Scholar search in October 2014 indicated that the 1980 Salter at al. JBJS publication has been cited approximately 1,096 times. Many of the articles that cite the 1980 JBJS study appropriately focus on the effect of CPM on either the histology of cartilage repair, or the effect of CPM on adhesions and joint stiffness.
However, Salter’s observation of decreased stiffness in animals treated with CPM has been extrapolated to clinical applications that were not included in his original work, most notably total knee arthroplasty (TKA).Today the clinical use of CPM after arthroplasty is controversial. A 2010 Cochrane review, for example, identified 20 randomized controlled trials of 1,335 patients in which CPM had been evaluated after TKA. The review concluded that there is evidence that CPM increases knee flexion range of motion, but “the effects are too small to be clinically worthwhile.” A more recent 2014 Cochrane review of 11 randomized clinical trials involving 808 patients concluded that there is not enough evidence to conclude that CPM reduces venous thromboembolism after total knee arthroplasty.
With respect to CPM after cartilage-repair procedures, many other investigators have confirmed the findings Salter reported in 1980 in animal models. Indeed, the basic-science support is strong enough that CPM has been commonly used in humans after cartilage repair, yet its actual efficacy in people remains controversial. For example, in a 2010 systematic review, Fazalare and co-workers reviewed 1,087 human clinical studies in which CPM had been used after cartilage repair procedures. In spite of that large number of studies, Fazalare was unable to find any randomized, controlled studies related to CPM, and heterogeneity among procedures and outcome measures in those articles precluded performing a meta-analysis.
Authors of today may be envious of the more than 6,900 words and 52 photographs, photomicrographs, and graphs (totaling 20 printed pages) that JBJS devoted to Salter et al. in 1980, and one can’t help but wonder what this classic JBJS paper would look like if modified to fit today’s standards. But the main message is this: in spite of high-quality basic science studies using animal models, there remains a need for well-controlled studies in humans to test the efficacy of CPM after cartilage repair and other procedures.
Thomas W. Bauer, MD, PhD
JBJS Deputy Editor for Research