Q&A: Clinical, Molecular Associations With Knee Injury Outcomes

In this Q&A, corresponding author, Fiona Watt, MBBS, PhD, FRCP, discusses her study that found the combination of effusion and hemarthrosis at the time of acute knee injury is associated with adverse outcomes at 2 years post-injury.

The combination of effusion and hemarthrosis at the time of acute knee injury is associated with adverse outcomes at 2 years post-injury, according to a study published in The Lancet Rheumatology.1 The study also found that measurable acute joint injury inflammatory response was associated with worse patient-reported outcomes, but to a lesser degree than these clinical factors.

In this Q&A, corresponding investigator, Fiona Watt, MBBS, PhD, FRCP, Centre for Inflammatory Disease, Imperial College London, and the Centre for Osteoarthritis Pathogenesis Versus Arthritis, University of Oxford, UK, discusses the research and its findings.

This longitudinal cohort study included 150 patients (82% male, median age 25 years) at a median of 17 days after a clinically significant acute knee injury. Researchers measured several predefined clinical variables at baseline, and 12 synovial fluid and 4 plasma or serum biomarkers at baseline and at 3 months. The primary outcome was Knee Injury and Osteoarthritis Outcome Score (KOOS4) at two years. Of the participants, 77 had all necessary variables available and were included in the analysis.

Make sure you never miss an article by signing up for our Rheumatology Network eNewsletter!

Rheumatology Network: Why was the trial conducted?

Fiona Watt, FRCP: We have been studying the links between tissue injury and osteoarthritis for a long time. This looks at this association at a molecular level in laboratory projects, but increasingly we wanted to test what we were finding in a “real life” situation, and this was when the KICK study was set up. Very simply, we wanted to test if the molecules in joint tissues that we found in pre-clinical models to be up-regulated by injury were also seen in the acute joint injury response in humans, and whether this immediate response predicted subsequent osteoarthritis development at 2 years after the injury.

RN: What were the surprises from the findings?

FW: As in other settings, it's probably no surprise that clinical factors when measured carefully might predict outcome as much as measurement of biomarkers, but that these can be complementary/additive. The presence of blood in the joint (not something we typically quantitate) as a strong predictor of outcome is perhaps not a surprise, but has not been described as a predictive factor in this setting before. That symptomatic outcomes and structural outcomes on X-ray had little association with each other, at least when we looked at the 2-year time point, was not something we were expecting and needs to be better understood. Again, this may not be a surprise for those familiar with the related literature in established osteoarthritis.

RN: How significant are the results?

FW: We desperately need better understanding of this whole area, both to understand osteoarthritis pathogenesis as a whole and also to try to predict and intervene specifically in post-traumatic osteoarthritis (which often affects individuals at a younger age). So, in developing our knowledge, the findings in that way are significant. For those trying to build a picture of important predictive factors which might feed into prognostic models. And those trying to find new therapeutic targets or mechanisms, I think this study generates lots of important avenues to pursue.

RN: What is the current practice and how could the findings possibly change things?

FW: It is important to state that the study was not designed to test current practice or to change it. However, our findings do throw a spotlight on the fact that many people with significant structural injuries and hemarthrosis are managed conservatively, sometimes with a lack of clear pathways of care. In the UK and increasingly globally, practice has moved away from arthroscopic or even arthrocentesis procedures. One question that this study’s findings pose is whether more could be done early, at the time of injury, and whether this might modify symptomatic outcomes and later disease risk. But this needs to be tested.

RN: What are the key takeaway points for clinicians who may be reading this article?

FW: Post-traumatic osteoarthritis, whether defined symptomatically or radiographically, was seen to be relatively common at 2 years after structural clinically significant knee injury in this group of predominantly young male sporting adults.

We found some predictors of worse outcome after acute knee injury. These were having a worse baseline Patient-Reported Outcome Measures (PROMs) score, the presence of moderate to large effusion or presence of substantial synovial fluid blood staining/ hemarthrosis at the time of the injury and an elevated inflammatory response measured in synovial fluid (specifically interleukin 6 [IL-6] and monocyte chemoattractant protein-1 [MCP-1] levels) at the time of the injury.

RN: Do you have anything else to add?

FW: These findings have not yet been replicated in other cohorts. We need to consider in our future studies and care of people with knee injuries whether an acute knee injury is a “joint attack” akin to a “heart attack;” an opportunity for early intervention aiming for prevention of future disease. Whether we can reliably identify those at higher risk in whom to intervene is an area of further work.

The work was funded by Versus Arthritis and Watt is funded by a UKRI Future Leaders Fellowship.


Garriga C, Goff M, Paterson E, et al. Clinical and molecular associations with outcomes at 2 years after acute knee injury: a longitudinal study in the Knee Injury Cohort at the Kennedy (KICK) [published online ahead of print, 2021 Jun 24]. Lancet Rheumatol. doi: https://doi.org/10.1016/S2665-9913(21)00116-8