Are Scoliosis Rods Permanent? And When Are They Removed?

Scoliosis rods are designed to be permanent. In the vast majority of spinal fusion surgeries for scoliosis, the metal rods and screws stay in your body for life, holding the corrected spine in place while bone graft fuses the vertebrae into a solid block. Removal is uncommon and almost always driven by a complication rather than a planned timeline. The situations that do lead to removal, and what happens to the spine afterward, are worth understanding in detail.

Why the Hardware Is Meant to Stay

The goal of posterior spinal fusion, the most common surgical approach for scoliosis, is to permanently straighten and stabilize a section of the spine. Surgeons attach rods to the vertebrae using pedicle screws, then pack bone graft material along the spine to encourage the vertebrae to grow together into one fused mass. The rods hold everything in the corrected position while that fusion matures, a process that typically takes six to twelve months. Once the bone has solidly fused, the rods are no longer doing much mechanical work. But there is no medical reason to go back in and take them out. A second surgery carries its own risks, and the hardware causes no harm in most people.

Modern instrumentation is remarkably durable. Dual-rod constructs with high-density pedicle screw placement can hold the spine firmly even in cases where fusion is incomplete or slow to develop, a condition called pseudarthrosis that can be difficult to detect because patients sometimes have no symptoms at all.¹ That mechanical reliability is part of why the default plan is to leave everything in place.

The Reasons Rods Do Get Removed

While routine removal is not standard practice, a number of specific complications can force the issue. Surgeons weigh the risks of a second procedure against the problem the hardware is causing, and in certain situations, removal is clearly the better option.

Late Infection

Infection is the most common reason for implant removal after scoliosis fusion. These are not the acute surgical-site infections that show up in the first few weeks; they are late-developing infections that can appear years after the original surgery. In a large series of instrumented scoliosis fusions, about 1.7% of patients developed a late infection an average of roughly three years after surgery.² A separate study found that the typical signs include wound drainage, local pain, swelling, and redness, with fever being relatively uncommon.³

The reason late infections are so stubborn is that bacteria can form a biofilm on the metal surface of the implants, essentially a protective colony that antibiotics alone cannot penetrate. A systematic review of postoperative spine infections found that while early infections can sometimes be treated with antibiotics and surgical washout while leaving the hardware in place, late infections generally favor instrumentation removal because the biofilm has had time to mature.⁴ Keeping the hardware in when a mature biofilm is present makes it very difficult to eradicate the infection.

Pain and Hardware Prominence

Some patients develop persistent, localized pain or tenderness over the hardware, especially where screws or rod connectors sit close to the skin surface. This is more common in thin patients and in areas where there is less soft tissue padding. A case series describing three adolescents who developed focal pain over prominent implants after posterior spinal fusion found that partial removal of only the symptomatic hardware provided good pain relief without any progression of the spinal deformity.⁵ Partial removal is appealing because it avoids the full scope of a complete hardware extraction while addressing the specific problem.

In a broader look at reoperations after primary scoliosis fusion, implant removal for pain or prominence accounted for about 15% of all revision surgeries, with some patients getting a complete removal and others getting only the offending pieces taken out.⁶

Metal Allergy

Metal hypersensitivity reactions to spinal implants are rare but documented. Patients who develop an allergic response to the nickel, chromium, or cobalt in their hardware can experience persistent skin reactions, local tissue inflammation, and pain that does not respond to typical treatments. In one reported case, a patient developed allergic eczematous dermatitis after thoracic spinal fusion, with skin biopsies showing the characteristic pattern of an allergic reaction and muscle tissue near the implants showing chronic inflammation with eosinophils, a type of immune cell associated with allergic responses.⁷ In these cases, removal of the hardware and replacement with a different material (or no replacement at all, if the fusion is solid) is the treatment.

Rod Fracture and Non-Union

Rods can break. When a rod fractures, it is often a sign that the bone fusion underneath never fully completed. The rod was carrying more load than it was designed to bear long-term, and eventually the metal fatigued. Non-union should be suspected when a patient has ongoing pain and imaging shows a rod fracture.⁸ The fix typically involves removing the broken hardware, re-establishing the fusion with new bone graft, and often placing new rods. This is not so much a “removal” as a “replacement” situation, though the original hardware does come out.

What Happens to the Spine After Removal

The biggest concern with taking out scoliosis rods is whether the curve will worsen once the metal scaffolding is gone. The answer depends heavily on whether the underlying bone fusion is solid and how long it has been since surgery.

One study that tracked patients after instrumentation removal found an immediate loss of about four degrees in the main thoracic curve and six degrees in the lower curve, with additional settling over time bringing the total loss to roughly ten degrees in the thoracic curve and nine degrees in the lower curve.⁹ Patients who had their hardware removed less than two years after the original fusion lost significantly more correction than those who waited longer, which makes sense since the fusion had less time to consolidate.

Older research on the Harrington rod era, when removal was more routine, showed a somewhat reassuring pattern: even in fused spines that lost some correction after rod removal, the curves tended to stabilize within the first six months.¹⁰ The spine finds a new equilibrium, and the loss does not keep progressing indefinitely. That said, there are documented cases where correction was lost rapidly after implant removal, and surgeons have emphasized that patients need to understand this risk and commit to follow-up imaging afterward.¹¹

One strategy to reduce correction loss when hardware must come out for infection is reinstrumentation: after removing the infected implants and clearing the infection, surgeons go back in and place new hardware. This approach has been shown to reduce the amount of correction that is lost compared to simply leaving the rods out.³

Growing Rods in Children Are a Different Story

Everything above applies to definitive spinal fusion, where the goal is a permanent construct. But in young children with early-onset scoliosis, the situation is fundamentally different. A child’s spine is still growing, and fusing it too early can stunt trunk height and lung development. Growing rods are a temporary solution designed to control the curve while the child gets taller.

Traditional growing rods require repeated surgeries every six to eight months to manually lengthen the rods and keep pace with growth. Magnetically controlled growing rods were developed to avoid all those trips to the operating room. These devices contain an internal magnet that a clinician can lengthen non-invasively using an external magnetic controller during an office visit.¹² Multicenter data on magnetically controlled growing rods has been gathered in both primary implantation patients and children converted from traditional growing rods.¹³

The intended endpoint for growing rods is conversion to a definitive spinal fusion once the child has reached sufficient skeletal maturity.¹⁴ At that point, the temporary growing rods are removed and replaced with permanent fusion instrumentation. So for this group of patients, rod removal is not just expected, it is the plan from the start.

Growing rod treatment is not without complications. Traditional instrumentation carries over a 50% risk of complications including deep wound infection, rod failure, and pull-out of the hardware from the bone.¹² Magnetically controlled rods reduce the number of surgeries but introduce their own issue: metallosis. Metal debris from wear at the rod junctions can accumulate in surrounding tissue, triggering chronic inflammation. Studies have found significantly elevated concentrations of titanium and other metals in tissue near the wear marks of these devices, with histological evidence of macrophage infiltration and fibrosis.¹⁵ Similar findings of chronic inflammation with foreign body granules have been reported in revision surgeries for growing rod systems.¹⁶

Rod Materials and How They Hold Up

The three main rod materials used in scoliosis surgery are titanium, cobalt chromium, and stainless steel. Each has trade-offs that affect both initial correction and long-term durability, though the choice does not usually change whether the hardware stays permanently.

Titanium is the most commonly used today, partly because it produces far less artifact on MRI and CT scans than stainless steel, which matters if a patient ever needs spinal imaging for another reason.¹⁷ However, titanium is a softer metal, and meta-analyses have found that titanium rods fracture significantly more often than cobalt chromium rods.¹⁸ Cobalt chromium is stiffer and more resistant to fatigue failure, which is particularly relevant in long fusion constructs or in adult deformity surgery where mechanical demands on the rods are greater.

Despite cobalt chromium’s durability advantage, comparisons of the two materials for adolescent scoliosis have found no significant difference in the final curve correction or the correction rate.¹⁹ In practical terms, both get you to the same place radiographically for typical adolescent curves. The difference matters more in complex adult cases and very long constructs where rod fracture becomes a greater concern.

Stainless steel rods, once the standard, have largely fallen out of favor for new surgeries. They are the strongest of the three materials, but they produce heavy MRI artifact and contain nickel, which raises the possibility of allergic sensitization. A comparison of complication rates across rod types found that stainless steel patients had the highest overall complication rate, while cobalt chromium patients had the highest rate of reoperation.²⁰

Patients Still Living with Harrington Rods

A sizable population of adults underwent scoliosis correction in the 1960s through 1980s with Harrington rods, a hook-and-rod system that predated modern pedicle screw fixation. Many of these patients are now in their 50s, 60s, and 70s, and their aging hardware can present new problems. Harrington constructs often fused long segments of the spine and did not account for natural spinal curvature in the side-to-side or front-to-back planes the way modern instrumentation does. Over decades, the spine above and below the fusion can degenerate, and the patient may develop pain, flatback syndrome (loss of the normal inward curve of the lower back), or progressive imbalance.

Revision surgery for patients with Harrington rods is complex. A study of 41 patients undergoing revision found high rates of complications, including temporary neurological deficits in about a fifth of patients and implant-related failures in roughly 15%, with nearly 20% eventually requiring further surgery.

One consequence of carrying long fusion hardware for decades is stress shielding. When rods bear a portion of the mechanical load that the vertebral bone would normally carry, the bone within the fused segment can gradually lose density. Research measuring vertebral bone density in patients with long-segment Harrington rod fusions found that bone density was significantly decreased in the middle of the construct compared to the ends, consistent with this stress-shielding effect.²¹ The weakened bone may complicate revision surgery if it ever becomes necessary, since screws need solid bone to grip.

Non-Fusion Approaches and the Question of Permanence

A newer technique called anterior vertebral body tethering avoids fusion entirely. Instead of rigid rods and screws, a flexible cord is attached to the outside of the curved spine through screws in the vertebral bodies. The tension on the cord slows growth on the convex side of the curve, allowing the concave side to catch up and gradually straighten the spine over time. This approach received FDA approval through a humanitarian device exemption in 2019 and is limited to skeletally immature patients with flexible curves.²²

The tether is intended to stay in the body, but it works differently from a fusion rod. Because the vertebrae are not fused together, the spine retains motion in the treated segment. The cord can break, and when it does, the clinical significance depends on whether enough correction has already been achieved through guided growth. Tethering is still a relatively new procedure with limited long-term data, and it is not appropriate for all scoliosis patients. Severe or rigid curves still require traditional fusion. But for the right candidates, it represents a fundamentally different answer to the permanence question: the hardware stays, but the spine keeps moving.

Imaging Considerations with Metal in Your Spine

One practical concern that comes up for anyone living with spinal hardware is what happens when you need an MRI. Modern titanium implants are MRI-safe, meaning you can go into the scanner without risk of the metal heating or moving. But the implants still distort the magnetic field enough to create blind spots on the images. Stainless steel produces the worst artifact, titanium the least, with cobalt chromium somewhere in between.¹⁷

For most people this never becomes an issue, but if you develop a new problem in or near your fused segment, the artifact can make it harder for radiologists to see what is going on. Specialized MRI sequences designed to reduce metal artifact have improved substantially in recent years and can help, though they do not eliminate the problem entirely. CT scans are generally less affected than MRI but still show some distortion around metal. Knowing what material your rods are made of is useful information to share with any doctor ordering imaging, since it helps the radiology team choose the right scanning protocol.