Spinal discs at risk, but not from the usual suspects

Smoking, drinking, insomnia, and being really out of shape are bad for your spine — probably worse than the usual suspects, like wear-and-tear or biomechanical glitches.

And we have a nice new citation for that.

Although this is clearly where the research has been leading for years, it has now been persuasively underlined by an exotic new study of the factors that contribute to spinal disc degeneration — lots of genetics and statistics. Guo and colleagues dug into several huge genome-wide association studies (GWAS), looking for genetic drivers of back pain … and really finding them.

The power and pitfalls of trawling the genomic ocean for answers

GWAS are a powerful way of sussing out how diseases are linked to genes and their associated traits and behaviours. GWAS are “fishing expeditions” for no fish in particular: by nature, they are unbiased, make no assumptions in advance about what they will find, and can unearth relevant genes no one even suspected. This is a modern marvel.

On the other hand, like industrial trawling, GWAS also have a bycatch: critters we do not want, signals that look more interesting than they are, false positives. There’s nothing necessarily unbiased about the analysis of this kind of data, and the sheer volume of number crunching means there’s plenty of room for statistical jiggery pokery. GWAS are powerful modern science that can show things we couldn’t dream of even twenty years ago, but they are also hard — and hard is also fallible.

Not that I suspect a problem here (but it’s a good thing to understand about big genetic studies in general).

Just correlations? No, not just that, that’s the cool part here

And so how Guo et al. dug into the GWAS is significant, and even more advanced: they used “bidirectional Mendelian randomization” (MR) … to get past the inevitable “just correlations” criticism.

MR can tease out causality, that tricksy holy grail of medical research. It cannot “prove” that X caused Y, but it can get us a fair bit closer. It can show if the arrow of causality is more likely to point one way, the other, or both.

A diagram of the intervertebral disc, with the inevitable illustration of an extrusion, making it a “slipped” disc. Medical illustrators never show us healthy discs. I wonder what those look like?

What genetics and statistics showed

So what “disease” were Guo et al. shining the GWAS and MR lights on? Intervertebral disc degeneration (IVDD). And they found some arrows pointing interesting directions.

Quick disc review: the intervertebral discs are pucks of tough fibrous tissue between vertebra. Although amazingly tough, they can slowly flatten, fray, dry, and even calcify. Arthritis of the discs, in short. Backs hurt for many reasons, but the decline of spinal anatomy is certainly in the mix. As discussed ad nauseam by therapy nerds, spinal problems seen on scans don’t always hurt, and there’s plenty of hurting without the visible signs — an important fact that is always in need of emphasis and elucidation.

What I like about this study is that it really helps to make more sense of that paradox. The degeneration is a bit of a bystander. The real show is overall health and fitness. This would have amazed me a few years ago. Now it’s more like, “Oh, yeah? More evidence of the obvious, eh? Nice.”

Guo et al. report that risk of intervertebral disc degeneration was 20% higher in smokers and drinkers (behaviours with strong genetic factors). If you think those disc risks are bad, it surges up to 80% with poor sleep, and 100% if you have hypertension. Sugary and fatty blood both also boost the hazard: 8-30% depending on the marker.

These were the factors with the clearest causality. For contrast, they also reported bidirectional causal relationships between back pain and hip osteoarthritis, cholesterol, and weight. In other words, it’s just as likely that having back pain leads to being heavier as the other way around.

Fitness clearly matters to backs — but not because of the core strength.

It’s always about the physiology

I have been banging this drum louder and more often with every year I do this job: it’s not the condition of our parts that determines what hurts, it’s our health that determines how much it hurts … and how likely it is to stop hurting. Fit folks get disc degeneration too! But probably less of it, less painful, and less persistently painful. And this new study harmonizes with this principle quite strongly. Guo et al.:

“This Bidirectional Mendelian randomization study provides evidence of complex causal associations between modifiable risk factors and IVDD. It is noteworthy that metabolic disturbances appear to have a more significant effect on IVDD than biomechanical alterations.”

As nicely summed up by Physio Meets Science:

“These findings contradict the widespread biomechanical paradigm that sees excessive physical strain as the primary causal factor for disc degeneration.”

Considering buying the PainSci guide to low back pain, which is like a compilation of twenty years worth of blog posts like this, a 200,000-word beast … but more organized than just a pile of posts. It’s written for both patients and professionals, like all my content. Read the large, free introduction.