PMMA in the Calf to Heal a Foot? An Interesting Twist on the Masquelet Technique #ActAgainstAmputation

Sometimes the most interesting part of a paper is not what the investigators implanted, but where they put it. Liu and colleagues have taken polymethylmethacrylate (PMMA)—the familiar bone cement behind the Masquelet induced-membrane technique—and moved it away from the diabetic foot wound. They implanted long chains of PMMA beads into tunnels along the medial calf, then removed them eight weeks later. The goal was ambitious: create a vascular, biologically active membrane at a distance and improve healing in the foot below.

The operation is not where you expect it

This was a retrospective comparative cohort of 138 people with Wagner grade 3–4 diabetic foot ulcers: 69 underwent the new operation and 69 received debridement plus negative-pressure wound therapy. The PMMA group received gentamicin-loaded cement molded into approximately 15–20 beads on a long steel wire. These bead chains were buried in medial muscular/subcutaneous tunnels in both lower legs.

But this was not a cement-only procedure. The surgeons also opened the deep posterior compartment and ankle canal and released the common peroneal nerve, tibial nerve, posterior tibial artery, and medial and lateral plantar neurovascular structures. Eight weeks later, the cement chains were removed in a second operation.

The authors report an extraordinary signal: 100% healing in the PMMA group, mean healing around 35 versus 49 days, recurrence of 8.7% versus 39.1%, fewer debridements, warmer dorsal-foot skin, and much less pain. Those numbers demand attention. They also demand unusually careful scrutiny.

Why the induced-membrane idea is biologically plausible

The Masquelet technique is established orthopedic biology. A PMMA spacer placed in a bone defect provokes formation of a vascularized pseudosynovial membrane. That membrane can express VEGF, TGF-β1, BMP-2, and other regenerative signals; after spacer removal, it serves as a protected chamber for bone graft.

The biology is not confined to bone. Animal studies have produced PMMA-induced membranes in subcutaneous and intramuscular pockets, including the thigh. We also have a growing diabetic-foot literature in which antibiotic-loaded PMMA is placed directly into the debrided wound or osteomyelitic defect to deliver local antibiotics, fill dead space, and generate a vascular bed for closure.

What appears different here is the attempt to turn that local foreign-body response into a remote therapeutic signal. I could find old PMMA bead chains in open tibial fractures, classic tibial Masquelet reconstruction, animal PMMA pockets, and several studies of cement directly in diabetic-foot wounds. I could not find an earlier human clinical report of long PMMA chains deliberately buried in otherwise closed calf tunnels to help a separate foot ulcer heal.

The neuromodulatory part may be the less interesting part

We have written before about physiologic “action at a distance”: spinal cord stimulation, remote ischemic conditioning, tibial transverse transport, periosteal distraction, focused ultrasound, and other interventions performed away from the wound that alter distal perfusion, inflammation, or neural signaling.

This paper belongs in that conversation—but, ironically, the nerve-decompression component may be the less novel part. Multilevel decompression of the common peroneal, tibial, tarsal-tunnel, and plantar nerves has been studied in diabetic neuropathy for decades. It can reduce pain and may change local microcirculation. That makes it clinically relevant, but scientifically inconvenient: the enormous improvement in pain and the increase in foot temperature could be explained by decompression rather than by a PMMA-induced angiogenic signal.

In other words, the paper combines two active treatments and compares them with neither. We cannot tell whether the cement, the decompression, their combination, or selection of patients produced the result.

Now for the cold water

This is an interesting study, but it is not yet a convincing one. It calls itself propensity-score matched without reporting the matching model, variables, caliper, discarded patients, or balance diagnostics. Baseline ulcer area, depth, location, arterial anatomy, WIfI stage, and revascularization history are missing. Those omissions matter more than small differences in age or HbA1c.

There are also internal reporting problems. The title includes Wagner grade 2, but no grade-2 patients appear in the table. Control healing time is reported differently in the text and table. A confidence interval for debridement frequency crosses zero despite P<0.001. Several baseline P values look one-sided even though the methods specify two-sided testing.

Most seriously, all 69 PMMA patients reportedly healed. Ordinary logistic regression with no failures in the treated group produces separation, yet the paper reports a conventional finite adjusted odds ratio without explaining use of a penalized or exact method. The control group’s actual healing percentage—the most important comparator—is not clearly reported.

The mechanism is asserted rather than demonstrated. Skin temperature is not a direct measurement of reconstructed microcirculation. The paper says membrane was biopsied for capillary assessment, but then acknowledges that no histologic or cytokine analysis was performed. ABI averaged approximately 0.49 while TcPO2 averaged approximately 60 mmHg, an unusual combination that receives no vascular explanation.

Other inconsistencies—gentamicin in Methods but tobramycin in Discussion, waived consent in one section but obtained consent in another, and follow-up variously described as 18, 36, and 48 months—further reduce confidence. Several references are also plainly mismatched to the claims they supposedly support.

What would make this persuasive?

The procedure also carries a substantial treatment burden: bilateral calf and ankle surgery, implantation of a permanent foreign material for eight weeks, and a second operation for removal. A report of no major cement-related or neurovascular complications in 69 patients is encouraging, but it requires independent confirmation.

• A prospective, component-controlled study separating decompression from remote PMMA implantation.
• Complete baseline wound, infection, offloading, vascular, and revascularization data.
• Direct perfusion measurements rather than skin temperature alone.
• Histology and quantitative angiogenic profiling of the induced membrane.
• Blinded healing adjudication, prespecified statistics that handle separation, and transparent reporting of every complication from two bilateral operations.
• Long-term ulcer-free, amputation-free, and reintervention-free survival—not merely initial closure.

Interesting enough to test—not ready to use

The most compelling part of this work is not that PMMA can generate a membrane; we have known that for years. It is the possibility that an induced membrane created away from a wound might participate in a limb-wide regenerative response.

That possibility is biologically intriguing and, as far as I can determine, clinically novel in this exact configuration. But novelty and validity are different things. For now, this belongs beside tibial transport and other “action at a distance” strategies as a provocative hypothesis—not as a practice-changing result.

Related reading

• What is the induced membrane technique for bone/wound healing and how can it help us #ActAgainstAmputation?
• Antibacterial Bone Cement—More Support for Use?
• Outcomes of integrated surgical wound treatment based on tibial transverse transport
• “Spooky Action at a Distance”: Physiologic Neuromodulation for Limb Preservation

References

1. Liu L, Gao Q, Zou D, et al. PMMA-induced membranous angiogenesis accelerates wound healing in refractory diabetic foot ulcers: A comparative cohort study. JPRAS Open. 2026;51:731–741. doi:10.1016/j.jpra.2026.07.033.
2. Masquelet AC, Fitoussi F, Begue T, Muller GP. Reconstruction of long bones by induced membrane and spongy autograft. Ann Chir Plast Esthet. 2000;45:346–353.
3. Pelissier P, Masquelet AC, Bareille R, Mathoulin-Pelissier S, Amedee J. Induced membranes secrete growth factors including vascular and osteoinductive factors and could stimulate bone regeneration. J Orthop Res. 2004;22:73–79.
4. Aho OM, Lehenkari P, Ristiniemi J, et al. The mechanism of action of induced membranes in bone repair. J Bone Joint Surg Am. 2013;95:597–604.
5. Viateau V, Guillemin G, Bousson V, et al. Histological characteristics of induced membranes in subcutaneous, intramuscular sites and bone defect. Orthop Traumatol Surg Res. 2013;99:823–828.
6. Liu C, You JX, Chen YX, et al. Effect of induced membrane formation followed by polymethylmethacrylate implantation on diabetic foot ulcer healing when revascularization is not feasible. J Diabetes Res. 2019;2019:2429136.
7. Mendame Ehya RE, Zhang H, Qi B, Yu A. Antibiotic-loaded bone cement for neuropathic diabetic foot ulcers complicated by osteomyelitis: a randomized controlled trial. J Diabetes Res. 2021;2021:9911072.
8. Chang W, et al. Outcomes of an integrated surgical wound-treatment mode based on tibial transverse transport for diabetic foot wounds. Front Surg. 2023;9:1051366.
9. Ostermann PA, Henry SL, Seligson D. Treatment of complicated tibial shaft fractures with the PMMA bead-pouch technique. Unfallchirurg. 1989;92:523–530.
10. Tu Y, et al. Lower extremity nerve decompression for diabetic peripheral neuropathy: a systematic review and meta-analysis. Plast Reconstr Surg Glob Open. 2022;10:e4478.

The signal is interesting. The study is fragile. The next experiment should separate the biology from the operation.