Could Metformin Help Musculoskeletal Inflammation?

One of the things I still enjoy most about being an Osteopath is that patients continue to make me think. After more than a decade in practice, I don't often come across conditions I've never seen before, but every so often somebody says something that sends me home with a completely different question running through my mind. It's one of the reasons I still spend evenings reading research papers. Clinical practice doesn't stop when the last patient leaves; sometimes that's when the interesting part begins.

Recently, I was treating a patient with a rotator cuff injury who returned for a follow-up appointment a few weeks after I had first seen her. As we talked through how she'd been getting on, she mentioned that she'd started taking metformin since our previous appointment. We hadn't discussed it before. She'd been experiencing irregular overnight blood sugar levels and, while looking into ways of managing them, had come across articles suggesting that metformin might also have anti-inflammatory effects. She spoke to her GP, who felt it was an appropriate medication to help manage her glucose levels, and so she started treatment.

The reason she mentioned it wasn't because she wanted my opinion on metformin. She simply said that, alongside her overnight blood sugar improving, her shoulder seemed to have settled much more quickly than she'd expected. She wasn't suggesting the medication had cured her rotator cuff injury, and neither of us were trying to draw conclusions from one person's experience. It was simply something she'd noticed. By the end of the appointment, however, I realised I was thinking about her observation far more than I'd expected to.

Since that appointment last week, I’ve found myself wondering whether there was actually any evidence behind what she'd experienced. Like most healthcare professionals, I'd always thought of metformin as one of the most widely prescribed medications for Type 2 diabetes. I knew how it worked in improving insulin sensitivity and helping control blood sugar, but if someone had asked me whether there was any credible research linking it with tendon healing or musculoskeletal inflammation, I honestly wouldn't have known how to answer. Curiosity got the better of me, so I started reading.

I expected to find a handful of speculative papers and not much else. Instead, I found an area of research that has quietly been developing for years. It's certainly not conclusive, mostly animal based trials and in places it's contradictory, but it is far more interesting than I anticipated.

What quickly became clear is that researchers didn't become interested in metformin because they believed it was a treatment for musculoskeletal inflammatory related pain. Their interest actually began with diabetes itself. We have known for many years that people living with Type 2 diabetes are more likely to develop musculoskeletal conditions such as frozen shoulder, rotator cuff disease, trigger finger and a range of tendon disorders. They are also more likely to experience slower healing after injury or surgery. Those associations have been recognised for some time, but understanding why they occur has become a major area of research.

Much of it comes back to the effect that persistently raised blood sugar has on connective tissue. Higher glucose levels promote the formation of advanced glycation end products, often shortened to AGEs, which bind to collagen and gradually alter its structure. Tendons become stiffer, less elastic and potentially less able to tolerate loading or repair themselves efficiently. At the same time, diabetes is associated with chronic low-grade inflammation, increased oxidative stress and changes in the small blood vessels that nourish our tissues. When you look at all of those factors together, it isn't particularly surprising that tendons may behave differently in someone with poorly controlled metabolic health.

Once researchers recognised that relationship, the next question seemed a logical one. If improving metabolic health creates a better environment for connective tissues, could a medication like metformin influence more than blood sugar alone?

That was where the research became particularly interesting.

Most of the evidence that generated excitement hasn't come from studies involving patients at all. It has come from laboratories. Researchers using cell cultures and animal models have repeatedly found that metformin appears to influence several biological pathways involved in inflammation, oxidative stress and cellular ageing. One pathway that appears repeatedly in the literature is the activation of AMP-activated protein kinase, better known as AMPK, which plays an important role in how our cells regulate energy and respond to stress. Laboratory studies suggest that activating these pathways may reduce inflammatory signalling and help protect cartilage and tendon cells from some of the damaging effects associated with ageing and diabetes.

Reading through those studies, it's easy to understand why researchers have become enthusiastic. They provide a biologically plausible explanation for why metformin could have wider effects than simply lowering blood sugar.

The important point, however, is that laboratory research is where ideas begin rather than where treatments are proven.

When I moved from the laboratory studies into research involving real patients, the picture became much less straightforward. Several observational studies involving people with Type 2 diabetes have suggested that those taking metformin may have lower rates of rotator cuff disease and some other musculoskeletal conditions than similar patients who are not taking the medication. There have also been studies suggesting lower levels of musculoskeletal pain among people prescribed metformin, although observational research can only identify associations rather than prove cause and effect.

More recently, researchers carried out a randomised controlled trial involving patients with Type 2 diabetes undergoing arthroscopic rotator cuff repair. The patients taking metformin developed less postoperative shoulder stiffness than those receiving placebo, which understandably attracted considerable interest. At first glance, that sounds like strong evidence supporting a beneficial effect.

Then, a much larger observational study using the TriNetX database analysed almost 3,000 patients taking metformin alongside a matched group who were not. Interestingly, the researchers reported higher rates of postoperative shoulder stiffness throughout the first year in the metformin group, although those same patients were less likely to require revision rotator cuff surgery. Suddenly the story wasn't nearly as simple as it had first appeared.

That, perhaps more than anything else, reminded me why it's so important to read beyond the headline of a single paper. One study rarely provides the full answer. Science usually advances by gradually building a body of evidence, with each study adding another piece to the puzzle. Sometimes those pieces fit together neatly and sometimes they appear to contradict one another before a clearer picture eventually emerges. At the moment, I think metformin and musculoskeletal health firmly sits in that second category.

By the time I'd finished reading, my thoughts had shifted quite a long way from where they'd started. I wasn't left believing that metformin should somehow become a treatment for shoulder injuries because the evidence simply isn't strong enough to support that conclusion. Equally, I wasn't left dismissing the research either. What struck me most was how often we separate musculoskeletal health from general health, when in reality the two are closely intertwined.

As Osteopaths, we spend a great deal of time talking about exercises, rehabilitation, movement, strength and manual treatment, all of which remain fundamental to recovery. Increasingly, though, research is reminding us that the environment in which those tissues are trying to heal matters just as much. Sleep quality, nutrition, blood sugar regulation, systemic inflammation, cardiovascular health and metabolic health all influence the body's ability to repair itself. Perhaps that's the bigger lesson emerging from this area of research, regardless of where metformin ultimately fits within it.

Looking back at my patient's experience, could metformin have contributed to the improvement in her shoulder? Possibly. Could improved blood sugar control have created a more favourable environment for healing? That's also possible. Equally, her shoulder may simply have improved because rotator cuff injuries often recover with time, appropriate rehabilitation, activity modification and osteopathic treatment. The honest answer is that we cannot separate those factors, and one patient's experience should never be used to prove that a treatment works.

What that conversation did achieve, however, was exactly what good patient conversations often do. It made me ask a question I hadn't previously asked, and that question led me to an area of research I'd probably have overlooked otherwise.

One of the principles of evidence-based practice is recognising that good clinical decisions don't come from research alone, nor from individual patient experiences alone. They come from bringing together the best available evidence, clinical experience and the individual circumstances of the person sitting in front of you. For me, this was a good reminder of that balance.

I suspect metformin will continue to be investigated over the coming years, particularly as researchers become increasingly interested in the relationship between metabolic health and musculoskeletal disease. Whether it ultimately proves to have a meaningful role beyond diabetes remains to be seen. For now, I think the evidence is best described as promising in places, conflicting in others and far from settled.

Sometimes patients leave you with more than a treatment plan to think about. Occasionally they leave you with a question, and every now and then that question turns into an evening spent reading papers, challenging assumptions and learning something new. I don't think my patient's observation changed the way I'll manage the next rotator cuff injury that walks into clinic, but it has certainly changed the questions I'll be asking, and I suspect that, in the long run, is just as valuable.

References

Lambova S. Pleiotropic Effects of Metformin in Osteoarthritis. Life (Basel). 2023;13(2):473.

Chang R, et al. Metformin use is associated with a lower risk of rotator cuff disease in patients with Type 2 diabetes mellitus. Diabetes & Metabolism. 2022.

Carvalho-e-Silva AP, et al. Association between metformin use and musculoskeletal pain in people with Type 2 diabetes. European Journal of Pain. 2021.

Feng J, et al. Metformin Reduces the Incidence of Shoulder Stiffness After Arthroscopic Rotator Cuff Repair in Patients With Type 2 Diabetes. American Journal of Sports Medicine. 2025.

Additional review of the TriNetX matched cohort study evaluating postoperative outcomes after rotator cuff repair in patients with Type 2 diabetes, demonstrating lower revision surgery rates but higher postoperative stiffness in metformin users.