As I continue to go through a long list of research studies on femoral acetabular impingement (FAI), I keep coming across studies that come to conclusions that are rather shaky. Take this study from the September 2008 Journal of Arthritis and Rheumatology (Nonspherical femoral head shape (pistol grip deformity), neck shaft angle, and risk of hip osteoarthritis: A case–control study) as an example.
This study investigates the relationship between pistol grip deformity (known also as “cam impingement” ) and arthritis. This is an important relationship to examine when thinking about the theory of FAI and how one develops arthritis because it provides theoretical justification for surgery.
Practically speaking, someone with hip pain is looking at the surgery to just get rid of the damn pain. But from the medical professional’s perspective, you also need to have a good understanding of what’s going on to really address the problem. That’s great. But that’s no guarantee that the current understanding of FAI and hip pain makes sense — particularly if it’s driven by studies that are too confident in their conclusions or by interpretations that place too much confidence in the conclusions. The current belief surrounding FAI and arthritis is that deformity in the shape of the bone causes premature wear-and-tear of cartilagnious tissue then bone…which leads to arthritis in the hip. So to prevent all that damage and to prevent arthritis, you need to identify the bony abnormalities early and “fix” them (read more here).
In this study in the Journal of Arthritis and Rheumatology, they looked at how one type of bony deformity, the pistol grip deformity, was related to arthritis. They had a pretty good sample size of people with signs of arthritis in x-rays and a pretty good size control group with no signs of arthritis (both groups were over 900 strong).
In the control group — that’s the one WITHOUT signs of arthritis, there were signs of pistol grip deformity in 3.61% of the total group. In the group with arthritis, the number was 24.27%. Now, that looks like a stunning, definitive number, doesn’t it? Here’s the conclusion:
“Our findings indicate that pistol grip deformity is associated with hip OA (osteoarthritis). The increased prevalence of pistol grip deformity and an abnormally low neck shaft angle in unaffected hips of cases with unilateral OA suggests that they are risk factors for development of hip OA. However, both a nonspherical head shape and an increase in neck shaft angle may occur as a consequence of OA.”
See that last sentence? Don’t skim it. Reread it, because it’s extremely important. The authors are suggesting that the bony deformities may actually be a result of arthritis rather than a causative factor. This is one alternative interpretation of the data, and it’s showing them being honest about the limit of their study. There is another option they didn’t consider, and that’s that the head shape and neck shaft angle are not factors at all, but that’s a different discussion.
The hypothesis is the pistol grip and the neck shaft angle are related to osteoarthritis. Since there is an observed higher prevalence of pistol grip deformity in the arthritis group, one might think there is some causative relationship between the pistol grip and the arthritis. Is that a sound conclusion? The authors are noting that it it’s not.
Let’s say you and I walk into a park where two parties are being held. In Group 1, everyone is laughing and smiling and slapping each other on the back. 19% of them are wearing gray t-shirts. 54% are wearing blue t-shirts. 27% are wearing red t-shirts. The rest are wearing t-shirts of varying colors and patterns.
In Group 2, the volume of celebration is decidedly less. There is less laughing, and some of the people look like they’re getting ready to leave. 65% of them are wearing gray t-shirts. 20% are wearing blue t-shirts. 15% are wearing red t-shirts.
Does this mean gray t-shirts are making the Group 2 party less fun? Or that the blue t-shirts are making Group 1 more fun?
Obviously not. In this example, there is an obvious absurdity in looking at t-shirts as a causative factor because our experience in social situations in real life informs us that the color of t-shirts at a party is hardly the deciding factor for how much fun one has. There other many other factors, and t-shirt color prevalence is very low on the list. We’ve all been to enough parties to have figured this out.
This is the old “correlation is not causation” rule. Just because you observe a difference in hips does not mean that the difference you observed is the causative factor.
Here is one more highlight in this study that should make you rethink the alleged causative role of the pistol grip deformity.
These researchers found that there was marked symmetry between the right and left hips on individuals in the asymptomatic group. They use this to come up with an interesting theory:
“…the striking symmetry between right and left hips of the measured features in normal controls and the apparent absence of an effect of age lend strong support to the hypothesis that these morphologic features are fixed constitutional characteristics of individuals. Therefore, in patients with unilateral hip OA (osteoarthritis), it seems reasonable to assume that the appearance of the unaffected hip is likely to represent the predisease appearance of the contralateral hip.“
So let’s translate that. We see symmetry in the asymptomatic population. Our statistical analysis shows that age is not a factor in the shape of one’s hips. From that, we think hip shape is fixed for life. Therefore, for people who have arthritis in one hip, the non-arthritic hip is what the arthritic hip looked like before it got bad.
I want to focus on two things here.
First, their statistical analysis does not show correlation between age and deformity in a group of people. Their conclusion is that hip shapes don’t change with age. Does that follow? Well, let’s go back to the park and look at the Group 1 party again. We run a statistical analysis, and there is no correlation between age of a person and the color of his/her shirt. Therefore the color of someone’s shirt never changes with age. Does that make sense? No.
The only way to determine whether hip shapes change with age is to follow multiple individuals over time and track for changes in hip shape (and the authors actually note this a few sentences later). I would also submit that looking at hip shape is largely barking up the wrong tree, but that’s something I already talked about in this video.
The second point to drive home is that no causation has been established from what they’ve found. The pistol grip deformity was not shown to be definitively linked to arthritis. If an individual has arthritis in one hip with the pistol grip deformity, and the other hip is fine but also has the pistol grip deformity, does it seem like the pistol grip deformity should be considered the deciding factor? Probably not. This is from the same study:
In controls, we found pistol grip deformity to be a predominantly bilateral feature that was present in ∼6% of men.”
This means that 6% of the men in the group WITHOUT arthritis had the pistol grip deformity on both hips. Suspicious, no?
What’s the takeaway here?
The science on FAI is nowhere near as definitive as it’s made out to be. Check out this blog post and this one for more on the shaky science. None of this is to say that someone’s pain isn’t very real and very frustrating (I’ve been there myself); my goal is to educate about the shaky rationale behind surgery to fix the problem.