Spinal Implant Failure

When the spine is fused, the surgeon places metalwork — typically titanium or cobalt-chrome — to hold the vertebrae rigidly together while a bone fusion forms between them. The construct usually consists of pedicle screws anchored into the vertebrae, rods running between those screws, set-screws or caps that lock the rods into the screw heads, and often an interbody cage placed in the disc space at the front. This metalwork is sometimes described as a permanent scaffold, but that description is misleading, and the misunderstanding sits at the heart of what implant failure actually is.

Spinal instrumentation is not a permanent load-bearing structure. It is a temporary internal splint. Its purpose is to hold the spine still long enough for living bone to bridge across the segment and fuse it solidly. Once that bony fusion has formed, the bone takes over the work of carrying load and shares stress with the metal, protecting it. The metalwork then becomes largely redundant, quietly retired inside a spine that no longer needs it.

The difficulty arises if the bone does not fuse. The metal is then left to carry the full cyclic load of every bend, twist and step, indefinitely and alone. No metal can do this forever; subjected to repeated stresses below its breaking point, it eventually fails through fatigue — the same process that snaps a paperclip bent back and forth a few dozen times. Implant failure and non-union are two different things that are nonetheless closely bound together: non-union is a failure of biology, implant failure is the mechanical consequence that tends to follow.

The most useful single way to think about spinal hardware is as a race between fusion and fatigue. On one side the bone is trying to fuse, a process that takes months. On the other the metal is accumulating microscopic fatigue damage with every load cycle. If the bone fuses first, the construct is protected, the fused bone shoulders the load and the hardware survives intact. If the metal fatigues first, the implant breaks — a rod fractures, a screw snaps, or a screw works loose. In the great majority of cases the breakage is not a sign that the metal was defective, but a sign that the fusion that should have relieved it never formed. This is why broken hardware is most often a signpost to an underlying non-union rather than a primary fault of the implant.

Pseudarthrosis, or non-union, is the failure of the bone to fuse — a failure of biology. Implant failure is the mechanical sequel; broken metal is usually the visible signpost of a non-union hidden beneath. Screw malposition is a screw placed in the wrong position at the original operation, a placement problem present from day one rather than loosening or fatigue developing over months. Adjacent segment disease is new wear at the mobile level next to a fused segment; the metalwork there typically remains intact. Infection deserves particular mention because bacteria can resorb bone around screws and loosen them without any mechanical cause, so a loose screw with no obvious mechanical explanation always raises the question of low-grade infection. Cage subsidence is the sinking of an interbody cage into the soft bone of the endplate, which is mechanistically about bone giving way rather than the device breaking and is therefore tied most strongly to bone quality.

Pedicle screws fail in a recognisable sequence. The earliest sign is loosening, in which a small amount of movement develops between the screw threads and the surrounding bone. This micromotion prevents new bone forming against the screw and the dead surface bone is replaced by a thin layer of fibrous tissue. On imaging this appears as a dark line, or halo, around the screw, and screw loosening is conventionally defined as a radiolucent zone of at least one millimetre — the halo sign. Once loosened, a screw can toggle within the bone, particularly at the top and bottom of a construct where leverage is highest, and may ultimately pull out altogether.

The dominant factor is bone quality, and specifically osteoporosis. The strength with which a screw holds is directly related to bone mineral density, and studies using opportunistic bone density measurement have found that patients with screw loosening have measurably lower density than those without. This is why the literature pays so much attention to bone health, and why techniques such as cement augmentation, in which bone cement is injected to anchor the screw, have been developed to improve screw purchase in soft bone, reportedly increasing pull-out force very substantially.

The rod is the long element connecting the screws and carries much of the bending load. In large series of complex deformity surgery, rods fracture in around one in fifteen patients overall, and the rate rises considerably after major corrective procedures such as a pedicle subtraction osteotomy. Rods do not break at random: they tend to fracture at or near the apex of a deformity correction or osteotomy site, where bending stress is greatest, and at the lumbosacral junction at the base of the spine, where loads are highest. Recognised risk factors include the presence of a non-union, a sharply contoured rod, the use of connectors or couplers, a construct spanning both the thoracolumbar and lumbosacral junctions, smaller-diameter rods, obesity and advanced age.

An early rod fracture, within the first year, tends to reflect primary mechanical overload before fusion has had a chance to take, and may itself delay union. A later fracture, after the first year, is more often the result of metal fatigue accumulating in a construct where the fusion never solidified. Either way, a fractured rod points back to the state of the underlying fusion.

An interbody cage is a small spacer placed in the cleared disc space to restore height and to hold bone graft while the two vertebrae fuse across the front of the spine. Cages can sink into the bone of the vertebra above or below — subsidence, conventionally defined as settling more than two millimetres into the endplate. They can slide out of position — migration — the most dangerous form of which is backward migration, or retropulsion, towards the spinal canal. Less commonly the cage itself breaks. Risk factors echo those for screw failure: osteoporosis, injury to the bony endplate at surgery, a single cage rather than two, a larger cage, and a cage positioned too far back. As with subsidence generally, the common theme is that these are problems of the bone giving way around the device as much as failures of the device itself.

A construct can also fail at the joints between its own components. Each rod is locked into each screw head by a small set-screw or cap, and longer constructs may join rods together with connectors or couplers, sometimes called dominoes. A loosened set-screw allows a rod to slide or migrate within the screw head, with or without the rod itself breaking; in rare reports a loosened rod has migrated a considerable distance. These junction failures represent a distinct failure mode that can occur even when the screws and rods themselves are sound, and the connectors used to join rods are themselves a recognised risk factor for rod fracture because they concentrate stress.

Broken hardware is not automatically a problem. If the spine has fused solidly, the bone is carrying the load and the metalwork has become redundant; a rod or screw that breaks in this setting may cause no symptoms and be discovered only by chance on a scan taken for another reason. Series of patients with rod fractures at the base of the spine have described entirely asymptomatic cases, all with a fusion confirmed on CT. The picture is different when the hardware fails and the spine has not fused: pain typically returns at the operated level, sometimes with a sensation of movement or a click, and in some cases gradual loss of the correction the operation achieved. In one notable series, only the symptomatic rod fractures came to surgery and every one of those was found to have a non-union, while the symptomless ones included cases with a solid fusion.

Assessment rests on imaging, and on a particular question that imaging is asked to answer: not merely whether the metal has broken, but whether the bone beneath it has fused. Plain X-rays are the first-line investigation; they show a frankly broken rod or screw, a migrated cage or the halo of a loosened screw, and bending films can reveal abnormal movement at a level that should be fused. They are relatively insensitive to early loosening and cannot reliably confirm a fusion. CT is the principal advanced investigation: it defines the fracture line through the metal, demonstrates the lucent halo around a loosened screw, shows the exact position of a cage, and, most importantly, is the best test for determining whether a solid bony fusion has actually formed. Peri-screw lucency on CT is regarded as the single best indicator of screw loosening. MRI is more limited because metal distorts the image, but with modern techniques is used to assess surrounding soft tissues, look for nerve compression, and, with contrast, investigate possible infection.

The evidence base is built on retrospective series, biomechanical studies and registry data rather than large randomised trials. Observation is the approach most often described for the broken implant causing no symptoms in a spine that has fused solidly: the fusion is doing the work, the metalwork is redundant, and across series the rate at which hardware failure leads to further surgery is low. Revision surgery is described where the failure is symptomatic, a non-union is confirmed or strongly suspected, deformity is progressing, or there is instability or pressure on the nerves. Techniques reported include replacing the broken rod, often with a stronger or larger-diameter one, adding extra rods to share the load, providing additional support across the front of the spine through an interbody fusion, revising and re-grafting the fusion bed itself, and, in osteoporotic bone, augmenting or upsizing screws and using cement to improve their hold. Revision for a broken implant is, in essence, revision for the non-union beneath it. Alongside surgery the literature consistently emphasises assessment and treatment of osteoporosis and cessation of smoking, both of which influence whether a fusion will succeed.

The outlook after implant failure depends almost entirely on the state of the fusion beneath the broken metal. Where the spine has fused and the failure is asymptomatic, the outlook is generally good; the broken hardware is a redundant passenger, the fusion is secure and intervention is frequently unnecessary. Where the failure reflects a non-union, the outlook is bound up with that non-union and recovery follows successful revision and the eventual achievement of a solid fusion. The consistent themes are that broken hardware should be interpreted in the light of the underlying fusion rather than in isolation, and that osteoporosis and smoking reward attention.

This page is general medical education and is not a substitute for individual advice. New or worsening weakness, numbness, saddle numbness or any disturbance of bladder or bowel control requires urgent assessment.

• —Sandén B, Olerud C, Petrén-Mallmin M, Larsson S. The significance of radiolucent zones surrounding pedicle screws. Definition of screw loosening in spinal instrumentation. J Bone Joint Surg Br. 2004;86(4):457–461. PMID 15174536.
• —Galbusera F, Volkheimer D, Reitmaier S, et al. Pedicle screw loosening: a clinically relevant complication? Eur Spine J. 2015;24(5):1005–1016. PMID 25616349.
• —Weiser L, Huber G, Sellenschloh K, et al. Insufficient stability of pedicle screws in osteoporotic vertebrae: biomechanical correlation of bone mineral density and pedicle screw fixation strength. Eur Spine J. 2017;26(11):2891–2897. PMID 28391382.
• —Bredow J, Boese CK, Werner CML, et al. Predictive validity of preoperative CT scans and the risk of pedicle screw loosening in spinal surgery. Arch Orthop Trauma Surg. 2016;136(8):1063–1067. PMID 27289259.
• —Löffler MT, Sollmann N, Burian E, et al. Opportunistic osteoporosis screening reveals low bone density in patients with screw loosening after lumbar semi-rigid instrumentation. Front Endocrinol. 2021;11:552719. PMID 33505353.
• —Chao KH, Lai YS, Chen WC, et al. Biomechanical analysis of different types of pedicle screw augmentation. Clin Biomech / Eur Spine J. 2015. PMID 25634339.
• —Smith JS, Shaffrey CI, Ames CP, et al. Assessment of symptomatic rod fracture after posterior instrumented fusion for adult spinal deformity. Neurosurgery. 2012;71(4):862–867. PMID 22989960.
• —Barton C, Noshchenko A, Patel V, et al. Risk factors for rod fracture after posterior correction of adult spinal deformity with osteotomy. Scoliosis Spinal Disord. 2015;10:30. PMID 26543505.
• —Hyun SJ, Lenke LG, Kim YC, et al. Comparison of standard 2-rod constructs to multiple-rod constructs for fixation across 3-column spinal osteotomies. Spine. 2014;39(22):1899–1904.
• —Dinizo M, Passias PG, Lafage R, et al. Pseudarthrosis and rod fracture rates after TLIF at the caudal levels of long constructs for adult spinal deformity surgery. World Neurosurg. 2021;155:e57–e63.
• —Yamato Y, Hasegawa T, Kobayashi S, et al. Rod fracture after apparently solid radiographic fusion in adult spinal deformity patients. World Neurosurg. 2018;120:e266–e271.
• —Aoki Y, Yamagata M, Ikeda Y, et al. A prospective randomized controlled study comparing TLIF techniques for degenerative spondylolisthesis: cage migration analysis. J Neurosurg Spine. 2012;17(2):153–159. PMID 22631523.
• —Kim MC, Chung HT, Cho JL, et al. Subsidence of polyetheretherketone cage after MIS-TLIF. J Spinal Disord Tech. 2013;26(2):87–92. PMID 22643181.
• —Zhao FD, Yang W, Shan Z, et al. Risk factors of cage subsidence in patients receiving MIS-TLIF. Front Surg. 2020. PMID 32472823.
• —Zavras AG, Federico V, Nolte MT, et al. Risk factors for subsidence following anterior lumbar interbody fusion. Global Spine J. 2024;14(1):257–264.
• —Aoki Y, Nakajima A, Takahashi H, et al. Cage positioning as a risk factor for posterior cage migration following TLIF — an analysis of 953 cases. BMC Musculoskelet Disord. 2015;16:213. PMID 26289079.
• —Chen CS, Chen WJ, Cheng CK, et al. Failure analysis of broken pedicle screws on spinal instrumentation. Med Eng Phys. 2005;27(6):487–496. PMID 15990065.
• —Kim YJ, Bridwell KH, Lenke LG, et al. Pseudarthrosis in adult spinal deformity following multisegmental instrumentation and arthrodesis. J Bone Joint Surg Am. 2006;88(4):721–728. PMID 16595461.
• —Cho W, Cho SK, Wu C. The biomechanics of pedicle screw-based instrumentation. J Bone Joint Surg Br. 2010;92(8):1061–1065. PMID 20675748.
• —Gertzbein SD, Robbins SE. Accuracy of pedicle screw placement in vivo. Spine. 1990;15(1):11–14. PMID 2326700.
• —Hilibrand AS, Robbins M. Adjacent segment degeneration and adjacent segment disease: the consequences of spinal fusion? Spine J. 2004;4(6 Suppl):190S–194S. PMID 15541666.
• —Rajaee SS, Bae HW, Kanim LEA, Delamarter RB. Spinal fusion in the United States: analysis of trends from 1998 to 2008. Spine. 2012;37(1):67–76. PMID 21311399.

This information is for general patient education and is not a substitute for an individual consultation. Always seek personalised advice from a qualified spinal specialist.