home  /  insights  /  construct-stiffness-working-length-and-stress-concentration
Biomechanical & Medical Device

The construct, not the component: working length, screw count and stress concentration

A plate that meets every applicable standard can still be part of a construct that fails. Working length, screw placement, junctions and mixed metals are assembly properties, not device properties.

July 30, 2026 · 7 min read

The short answer

A bone plate that conforms to every applicable specification can still fail, because the plate is manufactured, tested and released as a component but fails as part of a construct: the assembly of plate, screws, bone, gap and loading that only exists once the operation is over. That construct can be loaded well beyond what its geometry tolerates, and conformance testing will never reveal it. Working length, screw placement, junctions and mixed metals are properties of that assembly, not of the device as manufactured. The device and the construct are different objects governed by different evidence, and many disputes about orthopedic fixation hardware are really disputes about which one is being discussed.

What this article establishes

  • Whether an orthopedic fixation device conforms to its specification is a bounded, laboratory-answerable question, while the construct question is answered by geometry and loading from the operative record and the imaging, and evidence resolving one has almost no bearing on the other.
  • Working length, the unsupported span of plate between the screws nearest either side of the fracture, is the dominant variable in how hard a fixation implant is worked, but a very short working length produces a stiff construct that can suppress the interfragmentary motion callus formation depends on.
  • An empty screw hole over a fracture gap is at once a deliberate choice, since it lengthens working length, and the most common fatigue initiation site in fractured plates.
  • Whether a fixation construct shares load with bone or bears the whole load until healing is a property of the injury and the reduction, not of the implant, which is why an identical device can perform indefinitely in one case and reach the end of its fatigue life in another.
  • Contouring a plate, cutting a device to length and scratches from instrumentation create fresh stress raisers after manufacture. Dissimilar alloys introduce a galvanic couple, junctions with micromotion introduce fretting and crevice corrosion, and a corroded notch is an efficient fatigue initiation site.
  • An opinion that says which of material conformance, construct geometry and post-operative loading the evidence actually reaches, and which it does not, is more durable than one reporting a single conclusion about why the hardware broke.

Why are the device and the construct separate questions in orthopedic hardware failure?

The device and the construct are separate questions because they are answered by different evidence: the device question is answered by material and dimensional verification against the design specification and the applicable standards, while the construct question is answered by geometry and loading, from the operative record and the imaging rather than the metal. The standards that frame the device question are ASTM F138 and F139 for wrought stainless steel, ASTM F136 for titanium alloy, ASTM F382 for plates, ASTM F543 for metallic bone screws, ASTM F1264 for intramedullary devices, and ISO 5835 and ISO 5838 for screw and external fixator dimensions. The device question is a bounded, laboratory-answerable question.

Keeping the device question and the construct question separate is the first analytical step in analyzing an orthopedic fixation hardware failure, because evidence resolving one has almost no bearing on the other.

What is working length, and why does it matter for a bone plate?

Working length is the unsupported span of a bone plate between the screws nearest either side of the fracture, and it is the dominant variable in how hard the implant is worked. Bending stress rises steeply as that span increases, and the same plate can sit at very different points on its fatigue curve depending only on which holes were filled.

The relationship between working length and failure is not one-directional. A very short working length produces a stiff fixation construct that can suppress the interfragmentary motion callus formation depends on. A longer working length distributes strain but raises implant stress. Bridging constructs sit somewhere between those failure modes, and the choice of working length is visible on the first post-operative film.

Does the number of screws in a bone plate determine whether it holds?

Not by itself: screw count alone says little about a bone plate construct. What matters is the spread of fixation on each side of the fracture, purchase in each cortex, and whether the screws nearest the fracture were omitted.

An empty screw hole over the fracture gap is at once a deliberate choice, since it lengthens the working length of the plate, and the most common fatigue initiation site in fractured plates.

Bone screws fail by their own mechanisms: torsional overload during insertion, which ASTM F543 methods address directly, and bending fatigue at the head-shank junction or the thread run-out where the section changes.

What is the difference between load sharing and load bearing in fracture fixation?

A load-sharing fracture fixation construct has cortical contact restored across the fracture, so it shares load with bone from the outset and implant stress stays low; a load-bearing construct spans a comminuted or gapped fracture and bears the whole load until healing arrives, accumulating fatigue damage at full amplitude throughout. The distinction between load sharing and load bearing underlies almost everything else.

Whether a fixation construct shares load or bears it is a property of the injury and the reduction, not of the implant, and it is why an identical device can perform indefinitely in one case and reach the end of its fatigue life in another.

Where do fatigue cracks start in orthopedic fixation hardware?

Fatigue cracks in orthopedic fixation hardware start at geometric discontinuities. Screw holes, thread roots, the junction between a plate’s shaft and its broader head, and the cannulation of a screw all raise local stress above the nominal value. Manufacturers manage this through radii, surface finish and material selection, and the remaining stress concentration is an accepted design feature.

Damage introduced to fixation hardware after manufacture behaves the same way. Contouring a plate to fit anatomy, particularly bending through a hole, cutting a device to length, and scratches from instrumentation all create fresh stress raisers on surfaces finished to avoid them.

Why do mixed metals and modular junctions matter in fixation hardware failure?

Mixed metals and modular junctions matter because assembling components of dissimilar alloys — a stainless screw in a titanium alloy plate, or parts from different systems — introduces a galvanic couple, and any junction with micromotion introduces fretting and crevice corrosion regardless of alloy. Corrosion products generate a local tissue response and, more relevant mechanically, a corroded notch is an efficient fatigue initiation site.

Multi-component fixation constructs concentrate this corrosion at their interfaces: nail-to-locking-bolt, plate-to-locking-screw threads, and spinal rod-to-screw connections. Examining those interfaces at retrieval, with surface analysis where indicated, is what distinguishes junction-driven damage from bulk material behavior.

Do spinal instrumentation and external fixators fail as constructs too?

Yes: spinal instrumentation is the clearest case of construct-level behavior, and external fixators are similarly assembly-dominated. In spinal instrumentation, rod diameter and material, contouring, the number of levels spanned, and whether fusion occurred determine loading far more than any component’s own properties. ASTM F1717 exists precisely because spinal implants have to be tested as assembled constructs rather than as isolated parts, and rod fracture in a pseudarthrosis is the analog of plate fracture in a nonunion.

In an external fixator, frame geometry, pin diameter and spacing, and bar offset from the limb set the stiffness, and pin-track loosening changes that stiffness continuously during treatment.

What does the operative record need to supply for a construct-level analysis?

Construct-level analysis of fixation hardware is only possible if the assembly can be reconstructed, so the operative record has to supply implant lot and catalog numbers, the screw types and lengths in each hole, any intraoperative contouring or cutting, the reduction achieved, and the post-operative weight-bearing instruction. Implant logs, operating room notes and the immediate post-operative radiographs together usually supply it.

Where the operative record is thin, a construct-level analysis narrows to what the retrieved parts and the imaging independently support, which is a real limitation and should be stated as one.

How should an expert opinion on fixation hardware failure keep the device and the construct separate?

An expert opinion on fixation hardware failure keeps the device and the construct separate by saying which of material conformance, construct geometry and post-operative loading the evidence actually reaches, and which it does not. Such an opinion is more durable than one reporting a single conclusion about why the hardware broke.

Separating the device from the construct makes a fixation hardware failure analysis testable: material conformance is verifiable in a laboratory against a published specification, construct geometry is measurable from imaging, and post-operative loading is estimable from the clinical record within stated bounds.

This discussion of orthopedic fixation hardware is general technical orientation, not a failure analysis, an engineering opinion, or advice on any specific matter. Determining the cause of a particular incident requires hands-on examination by a credentialed expert.

For informational purposes only. Not engineering or legal advice, and not an opinion on the cause of any specific failure or on the conduct of any party.

Related

The practice area

failure-analysis assistanttriage · not a substitute for an expert
Happy to. Tell me what failed, how it failed, and whether the failed part and the scene are still preserved. That last one often decides what can still be established.