Injuries to the internal supporting structures of the knee (i.e., ligaments and menisci) have been shown increase the risk of knee OA, particularly in athletes with knee injuries early in life. Unrecognized degenerative damage to these structures may also predispose to an increased risk of knee OA. Here, Hill et al (Arthritis Rheum 52(3):794, 2005) examine the relationship of anterior and posterior cruciate ligament (ACL and PCL) rupture to knee pain and prevalent knee OA.
Methods:
Men > 45 years of age and women > 50 years of aged were assessed for self-reported knee pain and/or physician diagnosed knee arthritis. Patients were recruited from Veterans Administration (VA) Medical Centers and community sites in the Boston, Massachusetts area. Knee pain was confirmed by exam before enrollment. Knee radiographs were obtained in all subjects to assess for the presence of and severity of knee OA. Subjects with patellofemoral osteophytes were also classified as having knee OA (n = 72). MRI was performed on the more symptomatic knee (for patients with knee pain) or the dominant knee (those without knee pain) and assessed for integrity of the ACL and PCL by two radiologists blinded to patient knee symptomatology.
Results:
360 subjects, two-thirds of whom were male, with confirmed knee pain and radiographic evidence of knee OA were compared to 73 subjects without knee pain. Of the 73 subjects without knee pain, 48 (52% male) demonstrated evidence of radiographic knee OA while 25 (68% male) demonstrated no evidence of radiographic knee OA. 22.8% of subjects with knee pain and radiographic OA had complete ACL tears by MRI, while only 2.7% of controls without knee pain had complete ACL tears (p=0.0004). Complete PCL tears were uncommon, with only 0.6% of subjects with knee pain and radiographic OA having PCL tears compared with 0% of controls without knee pain.
Cases with OA and evidence of ACL tears on MRI tended to have more severe radiographic OA than those without ACL tears (see table below), including a higher prevalence of medial joint space narrowing. The prevalence of lateral joint space narrowing did not differ significantly between those with and without evidence of ACL tears on MRI.
| Cases (knee pain + radiographic OA) with ACL tear n= 82 |
Cases (knee pain + radiographic OA) without ACL tear n=276 |
p value | Controls (no knee pain) without ACL tear |
|
| Mearn age (years) | 68.4 | 66.4 | 0.50 | 65.3 – 66.8 |
| Mearn BMI kg/m2 | 31.7 | 31 | 1.00 | 28.5 – 29 |
| Severity of knee OA: K/L grade 0 K/L grade 1 K/L grade 2 K/L grade 3 K/L grade 4 |
4.9% 8.5% 19.5% 52.4% 14.6% |
23.2% 22.1% 37.7% 15.6% 1.4% |
<0.0001 | Median K/L grade in controls with radiographic OA (66%) = 0 |
| Medial joint space narrowing | 82.1% | 47.8% | <0.0001 | — |
| Lateral joint space narrowing | 16.4% | 11.3% | 0.28 | — |
| Pain (mm on visual analogue scale) | 44.3 | 44.1 | 0.95 | 0 |
| Previous knee injury | 47.9% | 25.9% | 0.003 | — |
Conclusions:
The prevalence of complete ACL tears is higher in people with knee pain and radiographic OA than in people without knee pain. Complete ACL tears are associated with more severe knee OA and a higher prevalence of medial tibiofemoral OA in people with knee pain and radiographic evidence of knee OA. Prevalent PCL rupture is uncommon.
Editorial Comment:
Although causality cannot be established from cross-sectional studies, these results are interesting in showing an association between ACL rupture and knee OA in people with radiographic knee OA. In particular, the high prevalence of unrecognized ACL tears is unexpected, but may be representative of the population studied (veterans), a group that may have a higher risk for traumatic knee injury than the general public. Nevertheless, the progression from ACL rupture to knee OA is deserving of further study in a longitudinally followed cohort. Particularly relevant to the study summarized here is how long it takes from ACL rupture until the development of symptomatic knee OA. In addition, the development of knee OA after acute full thickness ACL rupture may differ from that of chronic partial ACL tears that eventually rupture, particularly as the latter is often not amenable to surgical correction.
