What is the practical takeaway

In 2024 meta-analyses, the clearest way to sort these injections is by what job they are meant to do. Cortisone still makes sense when the aim is fast, temporary relief — especially during an inflammatory flare — because its benefit in knee osteoarthritis appears to be mainly short term, not durable. BMAC sits in a different category: it uses bone marrow aspirate concentrate rather than a blood draw, and some reviews report short- to mid-term improvement in pain and function, although results are inconsistent. Exosomes are the least settled option of the three.

None of these should be presented as a universal long-term fix for every knee with osteoarthritis. The comparison point that matters most in 2024 is often PRP: across 35 randomised trials, corticosteroids were broadly similar to hyaluronic acid, while PRP showed stronger mid- and long-term improvement than cortisone. By contrast, exosomes remain early-stage; a 2025 early human trial reported no adverse consequences and some improvement, but standardisation and regulatory problems still limit confidence. In practice, the choice usually comes down to three aims: quick symptom control, a more involved biologic-style option, or waiting for better evidence.

When cortisone still earns a place

A suddenly swollen, painful knee is one of the clearest situations in which cortisone still has a role. When osteoarthritis flares over a few days, with marked inflammation, disturbed sleep or a sharp drop in walking tolerance, the point of the injection is usually to calm symptoms quickly enough to restore basic mobility or create a short bridge to physiotherapy, travel, or another treatment step. That is a narrow but legitimate job, and it is different from trying to change the long-term course of the arthritis.

The time horizon matters. A 2024 meta-analysis of 11 randomised trials involving 842 patients found that intra-articular corticosteroids produced a clinically relevant benefit over placebo only at short-term follow-up; the effect was smaller by mid-term and absent at long-term review. In a separate 2024 meta-analysis covering 35 randomised trials and 3,348 patients, corticosteroids were broadly similar to hyaluronic acid overall, while PRP showed superior improvement at mid- and long-term follow-up, with clinically relevant differences favouring PRP at those later time points.

Repeated use is where the conversation becomes more cautious. Reviews of the literature raise concern about possible structural effects with repeated steroid injections, but the overall picture is not fully settled. Preclinical evidence is mixed rather than one-sided, with studies reporting both benefit and harm and no clear agreement on the best steroid, dose or protocol. For that reason, repeat cortisone is usually treated as an individual decision for a specific flare or circumstance, not as routine maintenance for knee osteoarthritis.

Why exosome injections are still early-stage

Exosomes are scientifically interesting because review literature describes them as a potential therapeutic avenue in osteoarthritis rather than as a simple joint lubricant or standard anti-inflammatory drug. Even so, a plausible mechanism in the lab is not the same as proven, repeatable benefit for people with knee osteoarthritis in routine practice.

The strongest reality check is regulatory and technical rather than theoretical. A 2025 review highlighted practical barriers that still matter: different groups use different isolation and characterisation methods, exosome content can vary between products, and the regulatory path remains unsettled. In practical terms, one clinic’s “exosome injection” may not be directly comparable with another’s.

Human knee osteoarthritis data remain sparse. A 2025 study of human umbilical cord mesenchymal stem cell-derived exosomes reported a randomised, double-blind, ascending-dose trial with no adverse consequences and some improvement in clinical scores and MRI findings. That is encouraging, but it is still early evidence. It does not yet settle how durable the effect may be, which dose or source is best, or which subgroup of knee OA patients is most likely to benefit. At this stage, exosomes fit better as a research-adjacent option than as a mainstream proven injection.

Who may consider BMAC and what recovery involves

BMAC is usually weighed up when simpler measures for knee osteoarthritis have stopped giving enough relief, but there is still a wish to stay on the joint-preservation side of the pathway rather than move straight to replacement surgery. It is a bone marrow aspirate concentrate procedure, and that matters practically because it is more involved than a simple blood-based injection. A 2025 narrative review found short- to mid-term improvement in pain and function in some studies, but also stressed that results remain inconsistent, so the ideal responder profile is still not firmly settled.

Recovery discussions are usually more involved than with a blood-draw biologic because there is both a marrow-harvest step and the knee itself to consider. In practical terms, the first phase is usually framed around soreness, stiffness and gradually rebuilding load rather than expecting an immediate change in symptoms. Exact rehab steps vary by clinician, knee severity and what else is going on in the joint, so activity is usually built back in stages rather than rushed.

The honest trade-off is that BMAC may appeal to selected patients who want a more involved biologic option and accept the extra procedure, but the evidence still does not support certainty about long-term results. The same 2025 review concluded that published outcomes are promising enough to justify discussion in some cases, especially earlier than obviously end-stage disease, yet not strong enough to define BMAC as a settled answer for every knee OA subgroup. In short, it is usually considered as an attempt to support the joint and delay bigger interventions, with improvement measured over months rather than days.

How BMAC differs from PRP

Set beside PRP, BMAC is better understood as a different trade-off rather than a more “advanced” upgrade. PRP is platelet-rich plasma, whereas BMAC uses bone marrow aspirate concentrate. That extra harvest step makes BMAC generally more involved than a straightforward blood-based PRP procedure.

The published comparison is not one-way. A 2025 narrative review described the wider literature as inconsistent: some analyses report greater clinical benefit with BMAC than PRP in selected knee osteoarthritis cohorts, but long-term superiority over PRP has not been firmly established. So, while BMAC may outperform PRP in certain analyses, the evidence does not yet support treating it as the default next step for everyone with knee OA.

That leaves PRP with an important practical advantage: its evidence base for knee OA is more mature overall. In a 2024 meta-analysis of 35 randomised trials involving 3,348 patients, PRP outperformed corticosteroids at mid- and long-term follow-up; and compared with still-early exosome products, it sits on firmer clinical ground as well. The real decision is usually less about “which is newer?” and more about the clinical goal: a simpler, better-studied blood-derived biologic, or a more complex marrow-derived option that may suit selected cases but brings extra procedural burden and less settled long-term comparative certainty.

What to ask before choosing any injection

The useful way to finish is with a consultation checklist rather than a sales line. For knee osteoarthritis, the decision is usually stronger when it is anchored to the job the injection is meant to do in that knee on that day.

• What is the goal here: settling a flare, creating a window for rehabilitation, or trying an option because simpler treatment has not helped?
• How long might benefit realistically last at this stage of knee OA, and what would count as success by 6 weeks or 3 months?
• What are the trade-offs: repeated corticosteroid exposure, the extra procedure involved with BMAC, or the still-preliminary status of exosomes?
• What evidence supports this choice for the severity seen on examination, X-ray or MRI, and for current function, swelling and alignment?

A recommendation tied to imaging, examination and objective functional assessment is usually more credible than one built mainly on marketing language. The broad pattern is now fairly clear: corticosteroid is mainly short-term, PRP and sometimes BMAC sit in the biologic-support discussion for selected knees, and exosomes remain early-stage; if that kind of consultant-led review would help, MSK Doctors accepts online booking without referral at mskdoctors.com.

1. [1] Corticosteroid injections for knee osteoarthritis offer clinical benefits similar to hyaluronic acid and lower than platelet-rich plasma: a systematic review and meta-analysis. (2024). https://doi.org/10.1530/EOR-23-0198 https://doi.org/10.1530/EOR-23-0198
2. [2] Intra-articular corticosteroid injections provide a clinically relevant benefit compared to placebo only at short-term follow-up in patients with knee osteoarthritis: A systematic review and meta-analysis. (2024). https://doi.org/10.1002/ksa.12057 https://doi.org/10.1002/ksa.12057