BMAC for Knee Arthritis:
What 10 Years of Evidence Actually Shows

Here is a number that should reframe how you think about knee replacement surgery: in one long-term study tracking subchondral BMAC injections for knee arthritis, only 20% of knees required total knee arthroplasty (TKA) over 15 years — compared to 42% in the intra-articular group.

That’s not a minor difference. That’s potentially thousands of patients avoiding surgery — or at least delaying it by a decade or more.

BMAC — bone marrow aspirate concentrate — has been used clinically for knee osteoarthritis for roughly 15 years, with escalating research activity over the past decade. The evidence is nuanced, more complex than either its advocates or critics acknowledge, and worth understanding carefully if you’re facing a knee arthritis diagnosis.

In our practice, we believe patients deserve the honest version of the data — not the marketing version and not the reflexive dismissal. Here is what we know, what we don’t, and what it means clinically.

What Is BMAC?

Bone marrow aspirate concentrate is exactly what it sounds like: bone marrow is harvested (typically from the posterior superior iliac spine — your lower back pelvic bone), centrifuged to concentrate the biologically active components, and then injected into the target joint.

What makes the concentrate valuable isn’t a single cell type — it’s the complexity of what BMAC contains:

• Mesenchymal stem cells (MSCs): Multipotent progenitor cells capable of differentiating into cartilage, bone, and fat tissue, and providing powerful paracrine (signaling) effects
• Platelets: Containing growth factors including PDGF, TGF-β, VEGF, and FGF
• Anti-inflammatory cytokines: Particularly IL-1Ra (interleukin-1 receptor antagonist), which directly counters the IL-1β-driven cartilage destruction of osteoarthritis
• Cytokines and chemokines that modulate macrophage polarization and synovial inflammation

MSCs in native bone marrow represent only about 0.001% of nucleated cells — which is why concentration is essential. After processing, BMAC contains a meaningful dose of these cells along with the growth factor milieu that makes them biologically active in a damaged joint environment.

The harvest procedure itself takes approximately 15–20 minutes under local anesthesia. Most patients describe it as pressure rather than significant pain. The entire same-day procedure — harvest, processing, and injection — typically takes under two hours.

The Science: What 10 Years of Evidence Shows

Short-to-Medium-Term Outcomes

A December 2024 systematic review in the Orthopaedic Journal of Sports Medicine analyzed 8 RCTs with a total of 937 patients comparing BMAC to other intra-articular injections. The findings were clinically instructive:

• BMAC demonstrated significant improvement in clinical scores vs. baseline starting at 1 month post-injection
• Pain scores at 6 months (P = .033) and 12 months (P = .011) significantly favored BMAC over hyaluronic acid (HA)
• No serious adverse events associated with BMAC were reported across all 8 trials
• The advantage over HA did not exceed the minimum clinically important difference (MCID) threshold — an important honest caveat worth understanding in context

A 2025 review in the Journal of Orthopaedic Surgery and Research synthesizing all published clinical trial data found that intra-articular administration of BMAC has shown effectiveness in clinical trials, with statistically significant improvements in VAS, WOMAC, and KOOS scores at 6, 12, and 24 months compared to baseline across multiple studies.

The honest takeaway from the short-to-medium-term data: BMAC consistently outperforms hyaluronic acid and shows a favorable safety profile. The effect sizes are real but not dramatic in most 12-month trial windows. The picture changes considerably when you look at longer time horizons — which most single-injection RCTs are not designed to capture.

The Long-Term TKA Avoidance Data

The most compelling long-term data comes from subchondral BMAC injection studies. A study with 15-year follow-up found that only 20% of knees in the subchondral injection group required TKA — compared to 42% in the intra-articular group. That is a clinically significant difference that has drawn considerable attention in orthobiologic research, because it speaks to the question patients actually care about: will I need a knee replacement?

A separate study in 30 patients with bilateral OA secondary to corticosteroid-induced osteonecrosis, randomized to TKA versus subchondral BMAC, found that at 12-year follow-up, Knee Score showed no significant difference between groups — but of the 30 knees treated with BMAC, only 3 required TKA. Those are remarkable numbers for a procedure that preserves the native joint and avoids the risks of major surgery.

The subchondral delivery route — injecting directly into the bone beneath the cartilage rather than into the joint space — may explain the long-term durability advantage. By targeting the subchondral bone, where the vascular supply and cellular environment can support regenerative activity, the treatment may address pathology at a level that intra-articular injection alone does not reach.

The MILES Trial — Honest Interpretation Required

The Multicenter Trial of Stem Cell Therapy for Osteoarthritis (MILES) — a phase 2/3 four-arm RCT at four sites with 480 patients, published in Nature Medicine in November 2023 — compared BMAC, adipose stromal vascular fraction (SVF), and umbilical cord tissue MSCs to corticosteroid injection (CSI).

The headline finding from the MILES trial: MSCs offered the same level of benefit as corticosteroids at 12 months. All groups improved. No group was dramatically superior.

Duke Health summarized the MILES findings clearly: MSCs are a safe treatment with no adverse reactions observed, but MSCs are not the “fountain of youth” they are often portrayed as.

Here is our honest interpretation of the MILES data: the trial used Kellgren-Lawrence grade II–IV patients (a wide heterogeneous range), a single injection protocol, and a 12-month follow-up window. MILES did not test the subchondral delivery route. It did not stratify by cell dose or quality — a critical variable given everything we know about concentration thresholds. And it did not follow patients long enough to capture TKA avoidance data, which is where the most compelling BMAC evidence resides.

What the MILES trial tells us: a single BMAC injection performs comparably to corticosteroid injection at 12 months in a heterogeneous OA population. What it does not tell us: that BMAC over a longer horizon, with optimized subchondral delivery and verified cell concentration, is equivalent to injecting saline. Those are different claims, and conflating them misrepresents the evidence in both directions.

Why Dosing Matters — The Cell Count Threshold

One of the most important signals in BMAC research is dose-dependence. Studies consistently show that outcomes correlate with MSC concentration. The threshold that appears clinically meaningful is approximately ≥6,500 MSCs/mL (or ≥2,000 colony-forming unit fibroblasts/mL in CFU-F assays).

Patients with concentrated preparations below this threshold respond less well — a finding from Pettine et al.’s pivotal 2015–2016 work showing that among patients who failed to improve and required surgery, 4 of 5 had MSC concentrations below 2,000 CFU-F/mL. This correlation between cell dose and outcome is not unique to orthobiologics; it mirrors what we see across regenerative medicine as a field.

The practical implication is significant: not all BMAC preparations are equivalent. Point-of-care concentration systems, harvest technique, processing equipment, and physician experience directly affect the quality and quantity of MSCs delivered. A subthreshold preparation injected into a KL-III knee is not the same intervention as a well-concentrated preparation delivered under imaging guidance. When BMAC studies fail to show benefit, the first question we ask is whether cell concentration was verified.

Anti-Inflammatory Mechanism

BMAC contains significantly higher levels of IL-1Ra than either leukocyte-rich PRP (LR-PRP) or leukocyte-poor PRP (LP-PRP). Since IL-1β drives the primary cartilage destruction pathway in osteoarthritis — by activating matrix metalloproteinases (MMPs) and suppressing chondrocyte function — the IL-1Ra advantage of BMAC may explain its more durable pain relief compared to other orthobiologic options.

A 2024 meta-analysis confirmed that BMAC showed the best pain relief at 12-month follow-up compared to other biologics. The anti-inflammatory mechanism is not merely adjunctive — it may be primary. In moderate OA, stopping the destruction cascade is at least as important as attempting to regenerate tissue.

The Gap in Standard Care

The standard algorithm for knee OA is: physical therapy → NSAIDs → cortisone injections → hyaluronic acid → total knee replacement.

Corticosteroid injections provide short-term pain relief (typically 6–12 weeks) but have demonstrated cartilage-damaging effects with repeated use. A 2019 RCT in JAMA (n=140) found that triamcinolone injections every 3 months for 2 years resulted in significantly greater cartilage volume loss than placebo injections. The treatment that most patients receive most frequently in the OA progression window is actively accelerating joint degeneration.

The standard algorithm also doesn’t account for the fact that roughly 20% of total knee arthroplasty patients are dissatisfied with their surgical outcome. TKA is a major intervention with real revision rates, infection risk, and prolonged recovery — and it’s often the only option presented once OA reaches a certain grade. The conversation that should happen — what can we do to make TKA unnecessary, or at minimum to delay it until you’re older? — rarely takes place in standard orthopedic care.

BMAC belongs in the conversation earlier. Particularly for patients under 65 with moderate OA (Kellgren-Lawrence II–III) who are not yet ready or appropriate for TKA, it represents a disease-modifying option with a favorable safety profile and meaningful long-term data — neither of which can be said for repeated corticosteroid injection.

The 361 HCT/P Regulatory Point

Under FDA regulations, autologous BMAC — using your own bone marrow, minimally manipulated — falls under the 361 HCT/P classification. This means it can be used clinically without the full Biologics License Application (BLA) approval process required for 351 products. This is why BMAC is available in clinical practice today. It is not an experimental treatment in the regulatory sense; it is a minimally manipulated autologous tissue product.

This is a meaningful distinction from exosomes and allogeneic cellular products — which are classified as 351 drugs and require FDA approval before clinical use. When you hear that “stem cell treatments aren’t FDA-approved,” the speaker is either referring to allogeneic products (true) or conflating two distinct regulatory categories (incorrect). Autologous BMAC operates under a well-defined and legally permissible regulatory framework.

How We Use This at Pravida Health

BMAC injection is the centerpiece of our CartiNova protocol for knee osteoarthritis. Our approach differs from a simple “stem cell injection” in several important ways that we believe directly affect outcomes.

1. Ultrasound-guided harvest and injection. All bone marrow aspiration and joint injections are performed under direct imaging guidance to optimize placement and minimize procedural risk. Blind injection — injecting without imaging confirmation — is not our standard.
2. Cell count verification. We assess MSC concentration at the time of processing. Based on the Pettine threshold data, you should not be injected with a subthreshold preparation. If the yield is inadequate, we address that before proceeding.
3. Patient selection. BMAC works best in specific patient profiles: Kellgren-Lawrence II–III OA, BMI under 40, active lifestyle, and no varus/valgus deformity greater than 10 degrees. We screen carefully because honest patient selection produces better outcomes — and because recommending a treatment to a poor candidate serves no one.
4. Combination protocols. In appropriate patients, we combine subchondral and intra-articular delivery, and may layer PRP or shockwave therapy to optimize the healing environment pre- or post-BMAC. The biology is more durable when the joint environment is prepared, not merely injected.
5. Follow-up and outcome tracking. We track clinical outcomes, pain scores, and functional status at defined intervals. Regenerative medicine requires the same rigor of outcome tracking that surgery does. If you’re not improving, we want to know early — not at year three. Schedule a consultation to discuss whether CartiNova BMAC is appropriate for your case.

What You Can Do Today

These five steps are actionable regardless of where you are in the OA progression spectrum — whether you’re newly diagnosed, managing moderate symptoms, or considering surgery. They represent the highest-yield, evidence-supported actions in advance of or alongside any regenerative intervention.

1. Get weight-bearing X-rays of your knee. Kellgren-Lawrence grading requires standing X-rays — not supine imaging. MRI is useful for soft tissue assessment, but K-L grading on weight-bearing films determines candidacy for BMAC and every other OA intervention. If you only have MRI images, you need standing films before any meaningful candidacy conversation.
2. Avoid repeated corticosteroid injections. If you’ve already had multiple cortisone shots, the evidence suggests you may be accelerating cartilage loss with each injection. Discuss alternatives with a regenerative medicine physician before your next shot — not after. The window for effective BMAC narrows as OA advances.
3. Optimize body composition before injection. Every pound of weight lost reduces joint load by approximately 4 pounds per step. Losing 10–15 pounds before BMAC meaningfully improves outcomes and is part of our pre-procedure protocol. This is not cosmetic advice — it is a direct mechanism effect on joint loading that predicts regenerative response.
4. Maintain and build quadriceps strength. Quadriceps weakness is one of the strongest modifiable predictors of OA progression. Even moderate resistance training — adapted to your current pain level — reduces OA progression rate and improves surgical outcomes if you do ultimately need TKA. Muscle is the joint’s best protection.
5. Get a second opinion before scheduling TKA. If you’ve been told you need a knee replacement and you’re under 65, see a regenerative medicine specialist first. Not because surgery is always wrong — but because it should be a fully informed decision, made after exploring the alternatives. Book a consultation at Pravida Health to discuss whether BMAC or our CartiNova protocol is appropriate before committing to a surgical path.

Frequently Asked Questions