Hardware Removal After Ankle Fracture Surgery
- sarangndesai
- 4 days ago
- 13 min read
═══════════════════════════════════════
HARDWARE REMOVAL AFTER ANKLE FRACTURE SURGERY
By Sarang Desai, DO — Fellowship-Trained Orthopedic Foot and Ankle Surgeon
Sports Medicine | McKinney and Flower Mound, Texas
═══════════════════════════════════════
┌─────────────────────────────────────┐
THE SHORT ANSWER
Ankle hardware is designed to stay in. It does not need to come out just because it's there.
But removal is common — 17% of patients within 3 years in a nationwide registry of 83,666 operatively treated ankle fractures, and 27% over a 20-year Finnish series.[1][2] The most frequent reason is not a mechanical failure. It's discomfort.[3]
The critical number patients are rarely given: hardware removal carries an overall complication rate of 9.6%, including wound problems, infection, nerve injury, and refracture.[4]
So the honest framing is: removal is an elective operation with real risk that reliably helps the right patient and reliably does nothing for the wrong one. Picking correctly is the entire job.
└─────────────────────────────────────┘
This is one of the most common second-opinion questions I see, and it's one where surgeons genuinely disagree with one another. Below is how I think through it.
───────────────────────────────────────
WHY THIS QUESTION EVEN EXISTS
Whether hardware should be routinely removed after the bone has healed — in the absence of significant hardware-related pain — is genuinely controversial, in part because few studies have addressed the question directly.[5]
Surgical textbooks differ in their recommendations. Some surgeons recommend removal for everyone. Others recommend it only for young, active patients. Many favor leaving implants alone entirely and are unconvinced of clinically significant adverse effects from retained metal. In practice, the decision is usually made by surgeon preference.[3][5]
┌─────────────────────────────────────┐
If two orthopedic surgeons give you different answers about your plate, neither is necessarily wrong. This is an area of real disagreement.[3][5]
└─────────────────────────────────────┘
That's exactly why it deserves a careful conversation rather than a default.
───────────────────────────────────────
WHY ANKLE HARDWARE BOTHERS PEOPLE MORE THAN HARDWARE ELSEWHERE
There's an anatomic reason ankle plates cause trouble at a higher rate than plates in the femur or forearm.
Internally fixed hardware in the ankle can cause pain, limit motion, and create skin protrusions or discomfort — because of the muscle deficiency in the region and the close anatomic proximity of skin to bone.[3]
Put simply: there's almost nothing between your fibular plate and your skin. No muscle padding. Just a thin layer of tissue, a boot rim, a ski boot, a work boot, a hockey skate.
The figures on this are striking. One study reported that 89.4% of internally fixed ankle fractures result in some report of discomfort — soreness over the implants and scar tissue, reduced ankle motion, or strain-related pain.[3]
Note carefully: some discomfort is nearly universal. That is very different from discomfort worth another operation.
───────────────────────────────────────
HOW OFTEN DOES HARDWARE ACTUALLY COME OUT?
Removal is becoming less common. In the 20-year Finnish data, the incidence of hardware removal fell from 31 per 100,000 person-years in 2001 to 13 per 100,000 in 2016 — with a particular decline in removals performed within the first 3 months. Annual costs fell from roughly €994,000 to €472,600.[2]
That trend reflects a real shift in thinking: fewer surgeons now remove hardware reflexively.
───────────────────────────────────────
WHO ENDS UP HAVING IT REMOVED — AND WHY
The Norwegian cohort of 997 patients is the most useful dataset here, because it separated removal for complaints from removal for infection. Of 170 patients who had hardware removed, 144 were for complaints and 26 for infection.[6]
Risk factors were completely different for the two groups:
For removal due to complaints:
- Female sex — male sex had HR 0.70 (p=0.047)
- Younger age — HR 0.79 per 10-year increase in age (p<0.001), meaning risk falls with age
- Syndesmosis screw — HR 0.70, i.e. lower risk of removal in this cohort (p=0.038)
- Longer operative time — HR 1.09 per additional 15 minutes (p=0.007)
For removal due to infection:
- Older age — HR 1.42 per 10-year increase (p=0.006)
- Current smoking — HR 3.15 (p=0.005)[6]
That smoking figure deserves emphasis. Current smokers had more than three times the risk of needing hardware removed because of infection. For anyone facing ankle fracture surgery in the first place, that's one of the most modifiable risks on the list.
The Finnish registry adds one more pattern: implant removal was more likely in younger patients and in bimalleolar or trimalleolar fractures.[1] More metal, more prominent metal, more active patient.
───────────────────────────────────────
DOES REMOVAL ACTUALLY MAKE PEOPLE BETTER?
This is the question that matters, and there is reasonable evidence on both sides.
Evidence that it helps — when the hardware is symptomatic.
A prospective study of 43 patients with symptomatic implants after ankle fracture ORIF, using the Short Musculoskeletal Function Assessment, found a statistically significant improvement in function at 6 months (Δ = −4.1; 95% CI −7.0 to −1.3; P=0.003). Both the bother index and the daily activities domain improved significantly.[7]
The authors' conclusion was measured but clear: there appears to be value in removing ankle implants in patients who report discomfort during daily activities.[7]
Notably, age, sex, BMI, smoking status, and ASA score had no significant effect on the degree of improvement.[7]
Evidence that it helps even with minimal pain.
A study of 80 cases undergoing routine removal after bony union — regardless of how much hardware-related pain they had — reported:[8]
- Pain score fell from 3.4 to 1.3 (0–6 scale)
- 72.5% experienced improved ankle stiffness
- 81.3% had less discomfort walking on uneven ground
- 80.8% were satisfied with having had it removed
Those authors concluded that routine removal could improve daily activities and satisfaction even when hardware-related pain is minimal.[8]
How I read these two studies together. They're both encouraging, and they're both limited — a 43-patient prospective series and an 80-case series, neither randomized against leaving the hardware alone. Pain improving from 3.4 to 1.3 is real, but so is the fact that pain scores drift downward with time regardless.[8] And an SMFA change of 4.1 points is statistically significant without being dramatic.[7]
┌─────────────────────────────────────┐
Removal helps symptomatic hardware. The evidence that it helps asymptomatic hardware is weaker, and the risk is the same either way.
└─────────────────────────────────────┘
───────────────────────────────────────
THE RISK NOBODY QUOTES: A 9.6% COMPLICATION RATE
This is the single most important number in this article, and most patients have never heard it.
The American Board of Orthopaedic Surgery examined 13,089 hardware-removal-only procedures performed over seven years. The overall complication rate was 9.6% (95% CI 9.1–10.1%) — 8.8% surgical and 1.5% medical/anesthetic.[4]
Complication | Rate (95% CI) |
Unexpected reoperation | 2.5% (2.2–2.7) |
Wound-healing delay or failure | 2.1% (1.8–2.3) |
Infection | 1.6% (1.4–1.8) |
Unexpected readmission | 1.6% (1.4–1.8) |
Continuing pain | 1.2% (1.0–1.4) |
Nerve injury | 0.6% (0.4–0.7) |
Bone fracture | 0.5% (0.4–0.6) |
Life-threatening complication | 0.4% (0.3–0.5) |
The authors' bottom line: although individual complications like infection, refracture, and nerve damage occur at relatively low rates, surgeons and patients should be aware that hardware removal carries a definite risk.[4]
Read the "continuing pain" line carefully. Roughly 1 in 80 patients has the operation and still hurts. That's the outcome I spend the most time preparing patients for, because it's the one that produces the most disappointment.
───────────────────────────────────────
SYNDESMOTIC SCREWS: A SEPARATE AND EVEN MORE CONTESTED QUESTION
About 20% of ankle fractures involve a syndesmotic injury requiring stabilization, usually with one or more screws.[9] For decades, routine removal at 3–4 months was standard teaching. The evidence has moved.
The systematic reviews say routine removal is not supported.
A 2016 systematic review found that both available randomized trials showed no difference in functional outcome between routine removal and retention, and all but one case-control series found equal or better outcomes when the screw was retained. The authors concluded there is little justification for routine removal — while candidly noting that all included studies had substantial methodological flaws.[9]
A 2024 systematic review of 27 studies reached the same place: 18 of 27 (67%) found no significant difference between routine removal and retention. Four retrospective studies (15%) suggested retention might produce worse outcomes, and two (7%) suggested removal introduced additional risk. No significant differences in ankle function were found among groups with different removal timings.[10]
A 2025 study of 330 patients categorized into no removal, routine removal, and removal for clinical indications found no significant difference in Olerud-Molander scores (median 85 vs. 80 vs. 80; p=0.646) — but complications were more frequent in reoperated patients (9% vs. 2%). Their recommendation: a selective, indication-based approach.[11]
But the disagreement is genuine. A separate retrospective study of nearly 300 cases found the mean AOFAS score was significantly better in patients who had the screw removed (92.6 vs. 88.4) — while also finding that loss of reduction was more frequent after removal (8.5% vs. 2.1%), that the tibiofibular clear space widened more after removal, and that removal patients had higher costs and more days of medical leave (21 vs. 15).[12]
Those authors concluded a strong recommendation either way cannot be given.[12]
┌─────────────────────────────────────┐
Where this leaves things: routine, calendar-driven syndesmotic screw removal is no longer supported. Removal for a specific problem — stiffness, prominence, pain, breakage — remains reasonable.[9][10][11]
└─────────────────────────────────────┘
One practical note: a broken syndesmotic screw is not an emergency and often causes no symptoms at all. Screws are meant to be temporary stabilizers, and many break as normal motion returns. One review specifically noted that intraosseous screw breakage may increase the need for removal due to pain — but breakage alone, without symptoms, is not an indication.[10]
More on this injury pattern: Ankle ORIF Surgery: What to Expect
───────────────────────────────────────
MY ACTUAL DECISION FRAMEWORK
Reasons I do recommend removal:
- Symptomatic prominent hardware — a lateral plate you can feel through the skin, that hurts in boots, skates, or dress shoes, or that's tender to touch every day. This is the best-supported indication.[7]
- Infection — a non-negotiable indication once the bone has united.[6]
- Skin breakdown or threatened skin over an implant.
- Mechanically failed hardware causing symptoms — a backed-out screw, a broken plate, migration.
- A screw penetrating the joint or blocking motion.
- Planned as part of another operation — if I'm going in for something else anyway, the marginal risk of removing hardware in the same setting is much lower.
Reasons I generally do not:
- "It's foreign, I want it out." Understandable, but not a medical indication, and the operation is not risk-free.[4]
- A broken screw with no symptoms.
- Airport metal detectors. This one comes up constantly. Modern implants rarely trigger them, and when they do, it's a 30-second conversation — not a reason for surgery.
- "To prevent arthritis." Retained hardware does not cause ankle arthritis. Cartilage injury and imperfect joint alignment at the time of the fracture do. If you have arthritis, that's a different discussion: Ankle Arthritis Treatment in DFW: What the 2026 AAOS Guidelines Mean for You
- Generalized, poorly localized ankle pain. This is the most important one, and it gets its own section below.
───────────────────────────────────────
THE MOST IMPORTANT TEST BEFORE AGREEING TO REMOVAL
If your ankle hurts, the question is not "should the hardware come out." The question is "is the hardware the reason it hurts?"
Those are not the same question, and confusing them is the single most common path to a disappointing operation.
Hardware pain usually:
- Is directly over the plate or screw heads — you can put one finger on it
- Is worse with direct pressure — boots, skate tongues, kneeling, crossing your legs
- Is superficial, not deep in the joint
- Improves when the pressure comes off
Pain that is probably NOT from hardware:
- Deep, inside-the-joint pain with weight-bearing — think cartilage injury or post-traumatic arthritis
- Catching, clicking, or locking — think loose body or cartilage lesion
- Stiffness without tenderness — think capsular contracture and loss of dorsiflexion
- Giving way or instability — think ligament problem
- Burning, shooting, or electric pain — think nerve irritation, which removal may not fix and can worsen
- Aching that moves around and is worse in the morning — think arthritis
The most useful diagnostic step in an unclear case is a targeted local anesthetic injection around the prominent hardware. If numbing that specific spot temporarily eliminates the pain, removal is far more likely to help. If it doesn't, removal is unlikely to be the answer.
The deep-joint pattern is worth taking seriously, because a cartilage injury sustained at the moment of the fracture is the most common reason an ankle still hurts long after the bone healed perfectly: Ankle Cartilage Injury (Osteochondral Lesion of the Talus)
And if the problem is stiffness rather than hardware, the answer is rehabilitation, not surgery: Physical Therapy After Ankle Fracture Surgery
───────────────────────────────────────
TIMING: WHEN CAN IT COME OUT?
The bone has to be solidly united first. Removing hardware from a fracture that hasn't fully healed risks refracture — and refracture occurred in 0.5% of removal procedures even in the general population.[4]
General guidance:
- Minimum ~6 months after the original surgery in most cases, confirmed radiographically
- Often 9–12 months for bimalleolar and trimalleolar fractures, older patients, and osteoporotic bone
- Syndesmotic screws, when there's an indication, are sometimes taken out earlier — around 3–4 months — though no timing interval has shown functional superiority[10]
- Longer if healing has been slow, if you smoke, or if you have diabetes
The decline in removals performed within the first 3 months in the Finnish registry reflects this more patient approach.[2]
───────────────────────────────────────
WHAT THE OPERATION IS LIKE
Straightforward, but not trivial.
- Outpatient, usually 30–60 minutes
- Often through the original incision, which is done partly for cosmetic reasons and partly because the tissue plane is already there
- Regional block plus sedation or general anesthesia
- Weight-bearing typically allowed fairly quickly, often in a boot for a few weeks, depending on how many screw holes were left in the bone
- Screw holes are stress risers. The bone is temporarily weaker where the screws came out. This is why impact activity and sport are held for roughly 6–12 weeks, and it's why the 0.5% refracture risk exists.[4]
- Full recovery is generally faster than the original surgery, but it is not a "quick in-and-out" with no consequences
Realistic expectation setting: most patients with genuinely symptomatic hardware are pleased — around 80% satisfaction in the published series.[8] But roughly 1 in 5 is not, and 1.2% report continuing pain as a recorded complication.[4][8]
───────────────────────────────────────
COMMON MISCONCEPTIONS
- "The plate has to come out eventually." No. Most ankle hardware stays in permanently without consequence.[9]
- "Metal sets off airport security." Rarely, and never a surgical indication.
- "A broken screw is an emergency." Usually asymptomatic and often expected, particularly with syndesmotic screws.[10]
- "Removing it will fix my stiffness." It may help some — 72.5% reported improved stiffness in one series — but stiffness after ankle fracture is usually capsular and rehabilitative, not hardware-related.[8]
- "It's a minor procedure with no risk." 9.6% overall complication rate.[4]
- "Removal will prevent arthritis." There is no evidence for this. Arthritis after ankle fracture comes from the injury itself.
- "Everyone with a plate has pain." Nearly 90% report some discomfort — most of it not significant enough to operate on.[3]
───────────────────────────────────────
IF YOU SMOKE, READ THIS
Current smoking carried a hazard ratio of 3.15 for needing hardware removed because of infection.[6]
This applies to nicotine in every form — cigarettes, vaping, pouches. If hardware removal is being planned, stopping nicotine beforehand is the highest-value thing you can do to change your odds, and it matters more than any technical detail of the operation.
───────────────────────────────────────
WHEN TO GET A SECOND OPINION
- You've been told your hardware must come out routinely, with no specific symptom driving it
- Your ankle hurts deep in the joint and removal is being offered as the solution
- You had syndesmotic screws and removal is scheduled purely by calendar
- Your ankle catches, clicks, or gives way — that's not a hardware problem
- You had removal already and the pain didn't change
───────────────────────────────────────
THE BOTTOM LINE
- Ankle hardware is designed to stay. Routine removal is not supported by the available evidence.[9][10]
- The usual reason is discomfort, not mechanical failure — and nearly 90% of patients report some discomfort, most of which doesn't warrant surgery.[3][6]
- Removal reliably helps genuinely symptomatic hardware, with significant functional improvement and around 80% satisfaction.[7][8]
- It carries a 9.6% overall complication rate, including 2.5% unexpected reoperation, 1.6% infection, 0.6% nerve injury, and 0.5% fracture.[4]
- Routine syndesmotic screw removal is no longer supported, though the literature genuinely disagrees and removal for a specific indication remains reasonable.[9][10][11][12]
- Smoking triples the risk of removal for infection.[6]
- The decisive question is whether the hardware is the source of the pain — not whether hardware is present.
- This is an elective operation. There is rarely a reason to rush it, and rarely harm in waiting.
───────────────────────────────────────
RELATED READING
───────────────────────────────────────
ABOUT THE AUTHOR
Dr. Sarang Desai is a fellowship-trained orthopedic foot and ankle surgeon with particular expertise in sports medicine. With over 15 years of experience, Dr. Desai serves as a professional sports team physician and treats the full spectrum of foot and ankle conditions — including ankle fractures, fracture-dislocations, complex and revision fracture surgery, symptomatic hardware, Achilles injuries, cartilage injuries, chronic ankle instability, total ankle replacement, and foot and ankle reconstruction.
He is a published researcher, an orthopedic implant inventor, a national lecturer, and a former University of Texas All-American athlete.
Offices in McKinney and Flower Mound, Texas, serving patients throughout Plano, Frisco, Denton, Lewisville, Southlake, Dallas, Fort Worth, Arlington, and the greater Dallas–Fort Worth metroplex and beyond.
SCHEDULE AN APPOINTMENT
📞 (972) 547-0047
📍 McKinney, TX | Flower Mound, TX
If you're wondering whether your ankle hardware should come out — or if your ankle still hurts after a fracture that healed — bring your X-rays and operative report to a focused evaluation. The goal is to find out what's actually causing the pain before deciding whether an operation will fix it.
───────────────────────────────────────
This article is for educational purposes only and does not constitute personal medical advice. Always consult a qualified physician about your individual situation.
References
The Diagnosis, Treatment, and Aftercare of Ankle Fractures. Baumbach SF, Polzer H, Ochman S, Mittlmeier T, Rammelt S. Deutsches Arzteblatt International. 2026;123(15):arztebl.m2026.0071. doi:10.3238/arztebl.m2026.0071.
Rehabilitation for Ankle Fractures in Adults. Lewis SR, Pritchard MW, Parker R, et al. The Cochrane Database of Systematic Reviews. 2024;9:CD005595. doi:10.1002/14651858.CD005595.pub4.
Minimum 5-Year Follow-Up Results: CROSSBAT (Combined Randomised and Observational Study of Surgery vs No Surgery for Type B Ankle Fracture Treatment). O'Keefe R, Naylor JM, Symes MJ, Harris IA, Mittal R. Foot & Ankle International. 2022;43(12):1517-1524. doi:10.1177/10711007221128562.
Surgery for Type B Ankle Fracture Treatment: A Combined Randomised and Observational Study (CROSSBAT). Mittal R, Harris IA, Adie S, Naylor JM, CROSSBAT Study Group. BMJ Open. 2017;7(3):e013298. doi:10.1136/bmjopen-2016-013298.
Surgical Versus Conservative Interventions for Treating Ankle Fractures in Adults. Donken CC, Al-Khateeb H, Verhofstad MH, van Laarhoven CJ. The Cochrane Database of Systematic Reviews. 2012;(8):CD008470. doi:10.1002/14651858.CD008470.pub2.
Concomitant Unstable and Stable Gravity Stress Tests on Weight-Bearing Stable Weber B Ankle Fractures Treated Nonoperatively: A 2-Year Outcome Study. Gregersen MG, Robinson HS, Molund M. The Journal of Bone and Joint Surgery. American Volume. 2023;105(18):1435-1441. doi:10.2106/JBJS.23.00195.
ACR Appropriateness Criteria® Acute Trauma to the Ankle. Smith SE, Chang EY, Ha AS, et al. Journal of the American College of Radiology : JACR. 2020;17(11S):S355-S366. doi:10.1016/j.jacr.2020.09.014.
Ankle Fractures in Diabetic Patients: A Critical Analysis. Goldberg EM, Polachek WS, Hynes K. JBJS Reviews. 2023;11(3):01874474-202303000-00003. doi:10.2106/JBJS.RVW.22.00147.
Nonoperative Versus Operative Treatment of Displaced Ankle Fractures in Diabetics. Lovy AJ, Dowdell J, Keswani A, et al. Foot & Ankle International. 2017;38(3):255-260. doi:10.1177/1071100716678796.
Special Considerations in the Management of Diabetic Ankle Fractures. Manway JM, Blazek CD, Burns PR. Current Reviews in Musculoskeletal Medicine. 2018;11(3):445-455. doi:10.1007/s12178-018-9508-x.
The Management of Ankle Fractures in Patients With Diabetes. Wukich DK, Kline AJ. The Journal of Bone and Joint Surgery. American Volume. 2008;90(7):1570-8. doi:10.2106/JBJS.G.01673.
Three Week Versus Six Week Immobilisation for Stable Weber B Type Ankle Fractures: Randomised, Multicentre, Non-Inferiority Clinical Trial. Kortekangas T, Haapasalo H, Flinkkilä T, et al. BMJ (Clinical Research Ed.). 2019;364:k5432. doi:10.1136/bmj.k5432.
Comments