Suture-Button vs Syndesmotic Screw

What Works

Ranked by how strong the evidence is. This card carries no exercise prescription, and that is deliberate: no study has ever shown that rehab should differ by implant, so inventing a program here would be dishonest. Your exercises come from your surgical team.

Cinematic rendering of the two fixation approaches across the joint between the lower leg bones

Tier 1 — Strong evidence

1. Load only to your surgeon's weight-bearing allowance, whichever implant you have HIGH

The landmark trial loaded both implants on an identical schedule: partial weight at 2 weeks, full weight at 6 weeks. No study has ever randomized the loading timeline by implant.

Evidence: STRONG for the absence of any implant-specific alternative (Andersen 2018, 97 patients, Level I). An implant-specific rehab protocol does not exist in the literature.

2. Read the implant as information about a second operation, not about how well your ankle was set HIGH

Reduction is identical: no difference in side-to-side alignment (p=0.42), joint gap (p=0.60), bone overlap (p=0.84), on 3D scanning (p=0.710), or across body weights. Hardware is where they split: breakage 0% vs 25-30%, removal 4-13% vs 22-80%, reoperation RR 0.21.

Evidence: STRONG. Concordant across every meta-analysis, the RCT-only pools, in vivo 3D CT, and the adolescent cohort.

3. Do not expect the implant to make the ankle feel better HIGH

An umbrella review of 19 research reviews benchmarked every reported difference against the smallest change a patient can actually detect. Most fell below it.

Evidence: STRONG (Nieuwenkamp 2026, umbrella review of 19 systematic reviews). One meta-analysis reached this verdict on its own effect.

Tier 2 and Tier 3 — moderate and emerging evidence

4. Expect screw breakage, and understand it MODERATE

Roughly a quarter to a third of screws break, and it is frequently painless. Average time to planned screw removal is 86 days, which lands squarely in the rehab window. Routine removal is no longer standard practice, so ask whether it is actually planned rather than assuming.

Evidence: STRONG for the rate, MODERATE for the framing.

5. In a high-demand athlete, the suture-button predicts an earlier return to sport MODERATE

14 weeks versus 19 weeks, and the implant was the only factor that reached significance. Use this to plan the runway, not to shorten the protection phase.

Evidence: MODERATE (Colcuc 2018, randomized, 54 patients, single centre).

6. The twisting plane is the blind spot EMERGING

A weight-bearing scan found outward twisting of the outer shin bone only in suture-button patients (p=0.004, 20 patients). A larger non-weight-bearing scan found twisting problems in 26% of cases spread evenly across both implants. They disagree on the culprit and agree on the lesson.

Evidence: EMERGING and contradicted. Hold loosely.

What Doesn't Work

Reading a p-value as a patient benefit. The single most common failure in how this literature gets cited. It persists because "grade A recommendation" is a quotable phrase and "below the smallest detectable difference" is not.
Speeding up loading because the note says "TightRope". Widespread belief, no trial behind it.
Treating a broken screw as a catastrophe. Common, usually painless, often just what a rigid screw does in a joint built to move.
Treating the suture-button as a guarantee. It has its own problem under load and slightly more wound issues, and it does not set the ankle any better.
Costing the suture-button with a 2015 removal rate. Its money case was built on a habit surgeons have since dropped.
Arguing the implant when the real question was upstream. Does this joint need fixing at all, and would repairing the inner ankle ligament have been better?

Red Flags

Call your surgeon if any of these happen.

Your ankle hurts AND feels like it is spreading or giving way when you stand on it. This is the one that matters most. In the landmark trial, this happened to 7 of 49 screw-fixed patients and 0 of 48 suture-button patients.
New instability under load — the ankle shifts or gives when you put weight through it.
A hot, red, leaking, or opening wound.
Pain getting worse week by week instead of settling.
A sudden pop with new pain and you cannot put weight on it like you could yesterday.

Refer to: the operating surgeon. Not your GP, and not the emergency department unless the wound is acutely infected or the ankle is grossly unstable. This is their implant and their reduction.

Not a red flag: a small click or pop around 2 to 3 months with no new injury, no new pain, and everything still working. Roughly a quarter of screws break, it is usually painless, and it is often just what a rigid screw does in a joint built to move. Worth mentioning at your next appointment. Not worth panicking about.

Cinematic anatomical rendering of the ankle mortise and the joint between the two lower leg bones

Return to Training

Clearance is a clinician decision made on what you can do, not on the date, and not on which implant you have.

Your surgeon has cleared full weight-bearing and progressive loading
No pain with widening, and no instability when you load it
Wound fully healed, no hardware symptoms needing surgical review
If implant removal was planned, it has happened or been formally deferred
Full, pain-free up-and-down ankle movement (no implant-driven difference is expected here)
Even single-leg loading tolerance, with no pain or swelling 24 hours later
Criterion-based clearance for cutting and pivoting from your treating clinician
You understand that a broken screw, if it happens, is not a re-injury

Right now, answer one question: how much weight has your surgeon actually allowed on that ankle, and until when?

If you cannot answer it, that is the call to make this week. It is a far better question than which implant you got, because that allowance is the only number that governs your recovery. Your implant does not raise it.

Takes less than 2 minutes. No equipment needed.

Conviction

MODERATE

Confidence is not uniform here, and the split is the entire point of this page.

Suture-button reduces breakage, removal, and reoperation	HIGH
Both implants restore identical alignment	HIGH
The score advantage is statistically real	HIGH
Suture-button reduces malreduction	MOD-HIGH
Faster return to sport	MODERATE
The score advantage is big enough to feel	LOW-MOD
Suture-button permits earlier weight-bearing	LOW
Suture-button is cost-effective	LOW
The implant should change your rehab	NO EVIDENCE

What would change my mind: "the implant doesn't change your rehab"

A multicentre trial of 200 or more adults randomized to both implant (suture-button vs screw) and loading schedule (full weight at 2 weeks vs 6 weeks), measuring function at 12 months against a pre-registered threshold for what patients can notice, with weight-bearing scan alignment as a safety endpoint. If faster loading proved safe with the cord but not the screw, the implant would finally mean something to your physical therapist. Nothing retrieved comes close to that design.

What would change my mind: "the function advantage isn't big enough to feel"

A single adequately powered trial of 300 or more patients (the field's entire randomized pool is about 490) with a pre-registered, patient-anchored threshold, measuring the proportion of patients who exceed it rather than an average difference. A 15% gap in responder proportion would establish a real advantage. What would not change it: a sixteenth meta-analysis of the same seven trials reporting p<0.05.

Go Deeper

Recovering from an injury and not sure which advice is real and which is just repeated? The Verdict breaks down one piece of evidence a week, in plain English, for free.

Join The Verdict — free

The Full Picture — Anatomy, Diagnosis & Evidence

What's Actually Going On

The joint just above your ankle, where your two lower leg bones meet, is not supposed to be rigid. Every time you bend your ankle upward, the wider front of the ankle bone pushes into the socket and drives the outer bone outward and backward by a fraction of a millimetre. That movement is normal, and every fixation decision is a position on whether to allow it.

A screw clamps the two bones together. In lab testing on donor ankles it actually squeezed the joint tighter than an uninjured one (7.9mm vs 9.1mm), and it survived more twisting force before failing than the cord did. A suture-button reproduced the normal gap almost exactly (8.8mm vs 9.1mm) while allowing the bones to rotate and slide.

Here is the part that gets left out. All of that is donor-tissue lab work, and it is the most over-quoted material in this entire subject. When researchers scanned 47 real post-operative patients in 3D, the two implants produced no difference in alignment at all (p=0.710 / 0.192 / 0.477). The lab finding did not survive contact with the clinic. The implant does not restore the alignment. The surgeon's reduction does, and then either device holds it.

Dark cinematic anatomical rendering of the lower leg bones and the ligament complex joining them above the ankle

How to Identify It

There are no bedside tests on this page, and that is a finding rather than a gap. No physical test can tell you which implant is in an ankle, and no examination can judge whether a surgical reduction was adequate. That is an imaging and surgical question, so the sensitivity and specificity numbers you would normally see here are genuinely unavailable by design.

What a clinician actually reads is the operation note:

Which implant, and how many
The weight-bearing allowance, and until when the only number that governs rehab
Was the inner ankle ligament repaired
Was the back of the ankle bone fixed
Is removal planned

What the imaging can and cannot see:

Plain X-ray settles side-to-side alignment, and is blind to twisting. It can look normal in an ankle that is badly set.
CT scan n=47 | no difference between implants found rotational problems in 26% of cases, spread evenly across both implants (p=0.234).
Weight-bearing CT n=20 | p=0.004 found outward twisting of the outer shin bone only in the suture-button group, and found that neither implant fully restored the joint compared with the uninjured side.

The two scanning studies disagree about the culprit and agree on the more durable lesson: twisting is the plane nobody can see well, and being badly set is the strongest known driver of a poor result.

Cinematic clinical rendering of ankle imaging planes and the tibiofibular relationship

The Debate

No clinical guideline covers this choice as of July 2026. No NICE, APTA, BOA, or JOSPT guideline on implant selection exists. The trials are ahead of the guideline. A formally graded review is registered but only its protocol has been published.

Is the suture-button functionally better? The central conflict

Shimozono 2019, meta-analysis of 5 trials

Function score 95.3 vs 86.7 (p<0.001). Authors award the suture-button a "grade A recommendation". The win is reported 18 times across 19 reviews.

VS

Nieuwenkamp 2026, umbrella review of 19 reviews

Benchmarked every difference against the smallest change a patient can detect. It fell below that threshold in 11 reviews for one score and 12 for the other.

Follow the umbrella review. Both are correct arithmetic; only one answers the patient's question. It is also the newest and the highest-altitude synthesis available.

Does the suture-button save money?

Neary 2017, cost-effectiveness

Suture-button is the dominant strategy: $1482 less per patient and higher quality of life. Modelled a 20% screw removal rate.

VS

Ramsey 2018, cost-effectiveness

Not cost-effective at baseline. Screws win if removal stays below 13.7%; the cord wins above 17.5%. Modelled modern practice.

Neither, cleanly. The entire disagreement is one assumption: whether screws get routinely removed. They no longer are, and Ramsey says it directly — abandoning routine removal "has changed the financial landscape". The suture-button's money case was built on a habit the field has dropped.

Does it let you load sooner? (The one that matters most to rehab)

Xie 2018, pooled analysis

Time to full weight-bearing significantly earlier with the suture-button (p=0.000).

VS

Andersen 2018, the landmark trial

Loaded both groups identically: partial weight at 2 weeks, full weight at 6 weeks.

Follow the trial. The pooled signal reflects surgeons doing different things across different studies, not a tested protocol. No study has ever randomized the loading timeline by implant.

Honest Limitations

1. The measuring stick is the weak link, and the whole debate rests on it

The research finding: nineteen reviews, the suture-button winning on function 18 times.

The real-world gap: most of those differences were smaller than the smallest change the score itself can detect. The main score is a mix of clinician-rated and patient-rated items whose threshold is itself disputed, which is why the umbrella review's own recommendation is to fix the thresholds before running any more comparisons.

The adjustment: never promise a patient a better-feeling ankle because of their implant. Tell them what it actually predicts: their odds of another operation.

2. Fifteen meta-analyses of seven trials is not fifteen pieces of evidence

The research finding: roughly 15 reviews, all pointing the same way.

The real-world gap: they re-pool substantially the same 5 to 7 trials (280-490 patients). The layer became self-referential, which is exactly why an umbrella review was needed. One pool finds no difference at 6 and 12 months where another finds significance at the same timepoints, which means the signal is unstable to which studies you include. An effect that appears and disappears with the study pool is not a robust effect.

The adjustment: weight the two trials and the umbrella review above the count of meta-analyses. Volume of agreement is not strength of evidence.

3. The practice it was built on no longer exists

The research finding: implant removal 40.2% for screws versus 3.7% for the cord.

The real-world gap: the screw's headline problem was removal, and the cord's case was built when routine removal was standard. It no longer is. A 40-80% removal rate describes a surgical era, not a property of the implant.

The adjustment: ask whether removal is actually planned for this patient, rather than assuming it from the implant.

4. It answers a question that was never the therapist's

The research finding: a large, consistent hardware advantage for the suture-button.

The real-world gap: no study randomized rehab content or loading timeline by implant. The one trial that specified loading gave both groups the same protocol.

The adjustment: the therapist inherits the implant and takes the ceiling from the surgeon. There is no implant-specific rehab to deliver.

The Nuance

Both arms here are surgery, so there is no surgery-versus-no-surgery choice on this page. What there is instead:

Where the suture-button genuinely wins (all hardware, none of it function): breakage 0% vs 25.4-30.9%. Removal 3.7-13% vs 22-80%. Reoperation RR 0.21 (95% CI 0.06-0.69). Being badly set 0.8-1.0% vs 11.5-12.6%. Recurrent widening 0 of 48 vs 7 of 49. Return to sport 14 vs 19 weeks.

Where it does not: identical reduction, identical movement, function mostly below what a patient can notice, an untested weight-bearing timeline, unresolved cost, slightly more wound issues, and possibly a twisting problem of its own.

When the screw is still the right call: resource-limited settings (higher implant cost is not associated with better patient-reported outcomes in ankle fracture surgery), a surgeon's established practice with good results, and anywhere symptomatic removal genuinely runs below about 13.7%, which flips the cost verdict.

The honest truth: the suture-button is a better implant and it does not give the patient a better ankle. Both statements are true, and the gap between them is the whole story. It breaks less, comes out less, and sends fewer people back to theatre, and those are real, large, repeatedly replicated advantages that matter to your year. But it does not set the ankle better, does not restore more movement, does not let you load sooner on any evidence anyone has actually generated, and the function advantage that 19 reviews have been reporting for a decade is usually smaller than the smallest difference a patient can detect.

And the questions that outrank all of it: in a Weber B fracture with a joint that reduces properly, outcomes were similar with and without fixation across 8 studies covering 292 fractures. Repairing the inner ankle ligament instead produced malreduction of 0-9% versus 20-35%, and removals of 5.8% versus 41%. An implant that was not needed cannot be the better implant.

Sources

Nieuwenkamp CL, Tausendfreund J, Schepers T, et al. (2026), Injury. Umbrella review of 19 systematic reviews, 2010-2025. Suture-button superiority reported 18 times; differences below the minimal clinically important difference in 11 reviews (OMAS) and 12 (AOFAS). The central paper. PMID 41638088
Andersen MR, Frihagen F, Hellund JC, et al. (2018), J Bone Joint Surg Am. Randomized trial, 97 patients, 2-year follow-up, Level I. Both groups partial weight-bearing at 2 weeks, full at 6 weeks. Recurrent widening 7 vs 0 (p=0.005). PMID 29298255
Migliorini F, Maffulli N, Cocconi F, et al. (2024), Arch Orthop Trauma Surg. Meta-analysis restricted to 7 Level-I trials, 490 patients. Screws: greater implant failure, removal, and malreduction. PMID 38740648
Shimozono Y, Hurley ET, Myerson CL, et al. (2019), Am J Sports Med. Meta-analysis of 5 randomized trials, 285 patients. Broken implants 0.0% vs 25.4% (p<0.001). PMID 30475639
Xu K, Zhang J, Zhang P, et al. (2021), J Foot Ankle Surg. Meta-analysis, 12 studies, 654 patients. No radiological difference on any measure. PMID 33518505
Onggo JR, Nambiar M, Phan K, et al. (2020), Foot Ankle Surg. Meta-analysis of 5 randomized trials, 280 patients. Authors state clinical significance of their own effect is "unlikely". PMID 30503612
McKenzie AC, Hesselholt KE, Larsen MS, et al. (2019), J Foot Ankle Surg. 6 studies, 275 patients. Reoperation RR 0.21 (95% CI 0.06-0.69). PMID 31474406
Zhang P, Liang Y, He J, et al. (2017), BMC Musculoskelet Disord. 10 studies, 390 patients. Implant failure 0% vs 30.9%; removal 3.7% vs 40.2%. PMID 28676078
Colcuc C, Blank M, Stein T, et al. (2018), Knee Surg Sports Traumatol Arthrosc. Randomized, Level I, 54 patients. Return to sport 14 vs 19 weeks (p=0.006). PMID 29224059
Ramsey DC, Friess DM (2018), J Orthop Trauma. Cost-effectiveness. Not cost-effective at baseline; screws win below 13.7% removal. PMID 29521685
Neary KC, Mormino MA, Wang H (2017), Am J Sports Med. Cost-effectiveness. Suture-button dominant, $1482 less. Directly contradicts Ramsey. PMID 27601151
Hennings R, Souleiman F, Heilemann M, et al. (2021), BMC Musculoskelet Disord. In vivo 3D CT, 47 patients. No alignment difference; rotational malreduction 26%, evenly distributed. PMID 34809628
Elghazy MA, Hagemeijer NC, Guss D, et al. (2021), Foot Ankle Surg. Weight-bearing CT, 20 patients. Outward fibular twisting uniquely associated with the suture-button (p=0.004). PMID 33422428
Lim B, Shaalan M, O'hEireamhoin S, et al. (2024), PLoS One. 8 studies, 292 Weber B fractures. Outcomes similar with and without syndesmotic fixation. PMID 38857233
Wang J, Stride D, Horner NS, et al. (2021), J Foot Ankle Surg. 9 studies, 508 patients. Deltoid repair: malreduction 0-9% vs 20-35%; removals 5.8% vs 41%. PMID 33218869

Full evidence base and citation validation: 35 sources, validator PASS 35/35 (32 anchored to a PubMed ID, 3 flagged as unverified landmark references).

The flexible implant breaks far less. It won't make your ankle feel better or speed up rehab.