eMedicine - General Principles of Internal Fixation : Article by Ronald Lakatos

You are in: eMedicine Specialties > Orthopedic Surgery > TRAUMA

General Principles of Internal Fixation

Article Last Updated: May 29, 2007

Fractures have been treated with immobilization, traction, amputation, and internal fixation throughout history. Immobilization by casting, bracing, or splinting a joint above and below the fracture was used for most long bone fractures, with the exception of the femur, for which traction was the mainstay of treatment. In the past, open fractures and ballistic wounds with long bone fractures were not amenable to standard fracture care because of the associated soft tissue injury and the difficulty in preventing sepsis; thus, they usually resulted in amputation, especially during the US Civil War.

Although the concept of internal fixation dates back to the mid 1800s, Lister introduced open reduction, internal fixation (ORIF) of patella fractures in the 1860s. Use of plates, screws, and wires was first documented in the 1880s and 1890s. Early surgical fixation initially was complicated by many obstacles, such as infection, poorly conceived implants and techniques, metal allergy, and a limited understanding of the biology and mechanics of fracture healing. During the 1950s, Danis and Muller began to define the principles and techniques of internal fixation. Over the past 40 years, advancements in biological and mechanical science have led to contemporary fixation theories and techniques.

Disruption of the endosteal and periosteal blood supply occurs with the initial trauma, and maintaining adequate blood supply to the fracture site is essential for healing. Hunter described the 4 classic stages of natural bone repair: inflammation, soft callus, hard callus, and remodeling. The inflammation stage begins soon after injury and appears clinically as swelling, pain, erythema, and heat. Disrupted local vascular supply at the injured site creates hematoma and prompts the migration of inflammatory cells, which stimulate angiogenesis and cell proliferation. After the initial inflammatory phase, the soft callus stage begins with an infiltration of fibrous tissue and chondroblasts surrounding the fracture site. The replacement of hematoma by this structural network adds stability to the fracture site.

Soft callus is then converted into rigid bone, the hard callus stage, by enchondral ossification and intramembranous bone formation. Once the fracture has united, the process of remodeling begins. Fibrous bone is eventually replaced by lamellar bone. Although this process has been called secondary bone union or indirect fracture repair, it is the natural and expected way fractures heal. Fractures with less than an anatomic reduction and less rigid fixation (ie, those with large gaps and low strain via external fixator, casting, and intramedullary [IM] nailing) heal with callous formation or secondary healing with progression through several different tissue types and eventual remodeling.

Anatomical reduction and absolute stabilization of a fracture by internal fixation alters the biology of fracture healing by diminishing strain (elongation force) on the healing tissue at the fracture site. Absolute stability with no fracture gap (eg, via ORIF using interfragmental compression and plating) presents a low strain and results in primary healing (cutting cone) without the production of callus. In this model, cutter heads of the osteons reach the fracture and cross it where bone-to-bone contact exists. This produces union by interdigitation of these newly formed osteons bridging the gap. The small gaps between fragments fill with membranous bone, which remodels into cortical bone as long as the strain applied to these tissues does not cause excessive disruption and fibrous tissue develops (nonunion). This method of bone healing is known as direct bone healing or primary bone union. Essentially, the process of bone remodeling allows bone to respond to the stresses to which it is exposed.

Based on the mechanical milieu of the fracture as dictated by the surgeon's choice of internal fixation and the fracture pattern, 2 patterns of stability can result that determine the type of bone healing that will occur. Absolute stability (ie, no motion between fracture fragments) results in direct or primary bone healing (remodeling). Relative stability (ie, a certain amount of fragment motion) heals with secondary or indirect bone union.

Kirschner wires (K-wires, 0.6-3.0 mm) and Steinmann pins (3-6 mm) have a variety of uses, from skeletal traction to provisional and definitive fracture fixation. Resistance to bending with K-wires is minimal, so they are usually supplemented with other stabilization methods when used for fracture fixation. Skeletal traction with K-wires is possible with the use of a K-wire tensioner, which, with application, stiffens the wire and allows it to resist bending load.

K-wires and Steinmann pins can provide provisional fixation for reconstruction of fractures while incurring minimal bone and soft tissue damage and leaving room for additional hardware placement. Planning pin placement is important to avoid the eventual permanent fixation devices, and, if possible, pins should be placed parallel to screws used for fracture compression. Depending on the diameter, pins may also be used as guide wires for cannulated screw fixation.

Permanent fixation options include fractures in which loading is minimal or protected with other stabilization devices such as external fixators, plates, and braces. Pin or wire fixation is often used for fractures of the phalanges, metacarpals, metatarsals, proximal humeri, and wrists. K-wires commonly supplement tension-band wire constructs at olecranon, patella, and medial malleolus fractures.

The K-wires have either diamond or trocar points that are simplistic in design and have limited ability to cut hard bone, a process that can lead to overheating. For this reason, they should be inserted slowly when power equipment is used, to avoid thermal necrosis. Image intensifiers are often used for optimal positioning of the fixation, especially with percutaneous insertion combined with closed reduction. The pins may have points at both ends, facilitating antegrade or retrograde fixation techniques; however, they are potentially dangerous because the unprotected sharp end may penetrate the surgeon, with possible inadvertent implantation of viral disease.

Steinmann pins are larger, may be threaded or unthreaded, and are currently used primarily for long bone traction in conjunction with a Böhler traction stirrup. Early techniques of fracture treatment consisting of pins for skeletal traction and incorporation into a cast were fraught with pin infections, loosening, and loss of reduction. This technique has been replaced with more advanced external fixation devices, internal fixation methods, and minimally invasive plating and IM devices.

Bone screws are a basic part of modern internal fixation. They can be used independently, or, in particular types of implants, they can be used together with the implant. The common design of a screw (see Images 1-2) consists of a tip, shaft, thread, and head. A round screw tip requires pretapping, whereas a fluted screw tip is self-tapping. The screw shaft is located between the head and the threaded portion of the screw. The screw thread is defined by its major or outside (thread diameter) and minor or root (shaft diameter) diameters, pitch, lead, and number of threads. The distance between adjacent threads is the pitch.

The lead is the distance a screw advances with a complete turn. Lead is the same as pitch if the screw is single threaded, and lead is twice the pitch if the screw is double threaded (faster screw insertion). The root diameter determines the screw's resistance to breakage (tensile strength). Screws are referred to by their outer thread diameters, bone type for intended use (cortical or cancellous, determined by pitch and major/minor diameters), and proportion of thread (partially or fully threaded).

Screw pullout strength can be affected by several factors. Bone composition (density) is the primary determinant of screw fixation. The total surface area of thread contact to bone (root area) is another factor in pullout resistance. Pretapping the screw hole theoretically reduces microfracture at the thread-bone interface but requires an extra step for insertion. Self-tapping screws have been shown to have no clinical difference from pretapped screws for fracture or plate fixation, eliminate the tapping step, and are now the industry standard. The fluted portion of the screw tip has less thread contact with the bone, so slight protrusion at the opposite cortex is recommended.

Pitch, the distance between adjacent threads, affects purchase strength in bone. Increasing the pitch increases bone material between the threads but decreases the number of threads per unit of distance.

The industry standard for the screw head is a hexagonal recess (see Image 2 ), which provides a large contact surface between the screw head and screwdriver and allows for optimal transmission of torque. A cross-type screw head is used on some screws in the 2.0 and smaller screw (minifragment) sets (see Image 2). The star design found in industry has been used in the Alta system and screw heads for the Association for the Study of Internal Fixation (AO/ASIF) locking plates and has been shown to be superior for torque and resistance to stripping (see Image 2).

Several forces are involved with screw insertion and tightening. Torque is applied through the screwdriver to the screw head in a clockwise rotation to advance the screw in the predrilled path or, in the case of a cannulated screw, over a guide wire; this advancement produces a circumferential force along the thread. For cortical screws, the drill diameter is slightly larger than the root (shaft) diameter of the screw. Axial tension is created with impingement of the screw head on the cortex or plate, generating tension through the screw. To optimize these forces, screws should ideally be inserted at 80% of the torque needed to cause them to strip. An estimated 2500-3000 newtons of axial compression force can be applied to the average screw. Over time, the amount of compressive force decreases slowly as the living bone remodels to the stress; however, the fracture healing time is usually shorter than the time it takes for substantial loss of compression and fixation.

The 2 basic types of screws available for the variability of bone density are cortical and cancellous screws. Cortical screws are designed for compact diaphyseal bone, whereas cancellous screws are designed for the more trabecular metaphyseal bone. Cortical screws have a smaller major (thread) diameter, decreased pitch, and a shallower thread than cancellous screws. Standard nonlocking cortical screw diameter choices include 1.5, 2.0, 2.7, 3.5, and 4.5 mm.

Cancellous screws typically have a larger major (thread) diameter and pitch and a greater difference between major and minor (shaft) diameters in comparison to cortical screws, providing more surface area for bone purchase. These screws are intended for use in metaphyseal fixation, where bone is softer. Cancellous screws are available in sizes 4.0 and 6.5 mm, and variety of cannulated sizes from 4.0-7.5 mm.

Tapping is not usually necessary with cancellous screws in metaphyseal bone because cancellous bone is porous, and, in general, tapping is avoided with either type of screw because screw insertion compresses the cancellous bone, which increases the local bone density in contact with the thread and potentially increases screw purchase. Occasionally, tapping of cancellous screws is necessary in young, strong metaphyseal bone. If the cancellous screw does not advance easily in this situation, then tapping is required.

Positional or neutralization screws are to attach an implant, such as a plate, to bone by compression between the plate and bone (see Image 3). This function is modified when the screw is used to lag across a fracture through the plate or when used for fracture compression as with a dynamic compression screw. For a positional screw, the pilot hole is drilled with the appropriate-size bit (shaft diameter) for the screw to be inserted (eg, a 3.2-mm drill bit for a 4.5-mm screw) using a centering guide for the plate hole. A depth gauge is used to determine appropriate screw length, and the thread cut is then made with an appropriate tap or without a tap when self-tapping screws are used or screws are placed in metaphyseal bone.

Interfragmentary lag screws provide compression across 2 bone surfaces using the lag technique. A lag screw is a form of static compression and is applicable to intra-articular fractures to maintain reduction and diaphyseal fractures for stability and alignment. Ideally, lag screw fixation (see Images 3-4) produces maximum interfragmentary compression when the screw is placed perpendicular to the fracture line. Most lag fixation techniques require additional stabilization to neutralize the axial bending and rotational forces applied to the bone during functional postoperative care. This is provided by a neutralization or buttress plate or external fixation.

If lag screws are to be used without neutralization plate fixation, especially in long spiral fractures (>2 times the diameter of the involved bone), the ideal inclination of the screw is half way between the perpendiculars to the fracture plane and to the long axis of the bone. Placing the screw perpendicular to the long axis of the bone can also be considered because longitudinal or shear compression may cause the screw or screws to tighten. Interfragmentary screw fixation alone may be appropriate for avulsion injuries in which shear forces generate metaphyseal and epiphyseal intra-articular fractures, provided bone quality is good.

A fully threaded screw can serve as a lag screw (see Image 5) with the near cortex over-drilled to the size of the screw's major (thread) diameter (4.5 mm in the example). Once the near cortex is drilled, which provides a gliding hole, a drill sleeve with the outer diameter of the drill bit (4.5 mm) is inserted into the hole and the standard drill bit (3.2 mm shaft diameter) is used to drill the far cortex. As the screw threads grasp the distal cortex, compressive forces are generated through the axis of the screw to the screw head, causing the 2 fracture fragments to be compressed. This same mechanical effect can be generated by a partially threaded screw, with all threads entirely within the opposite bony fragment.

Cannulated screws are now provided by most trauma manufactures in sizes from minifragment to 7.5 mm, usually with a cancellous thread, but cortical patterns are also available. They are inserted over a guide wire usually placed under fluoroscopic control, which allows for initial provisional fixation to be incorporated with definitive fixation. Cannulated screws allow for a percutaneous technique, such as is used with hip pinning, or may be used with limited open reduction techniques and can help minimize soft tissue dissection and periosteal stripping. Most designs are now self-drilling and self-tapping, but some may require predrilling over the guide wire, especially in dense bone. The guide wires are usually terminally threaded, although nonthreaded are available, and, when drilling over the wire, the author recommends that the drill be stopped before crossing the threaded portion or the guide wire will be removed with removal of the drill bit. This could result in difficulty relocating the drill hole,and provisional fixation may be lost.

The pullout strength of cannulated 7-mm cancellous screws versus 7-mm noncannulated screws and 3.5-mm cannulated and noncannulated screws has been tested in 2 studies, and no significant difference was noted regarding pullout strength. However, these studies are specific to these screw designs, and similar fixation properties cannot necessarily be applied to other screw designs and sizes. Also, keep in mind that the relative costs of these implants are often 10 times those of similar-sized noncannulated screws, which should be used when appropriate.

Self-tapping screws have the advantage of eliminating a step during screw insertion, thereby decreasing operative time. The fluted design of the screw cuts a sharp path in the predrilled hole, eliminating the need for tapping. These screws have not been recommended as lag screws because of the tendency for repeated reinsertion if the thread hole cannot be found. In this situation, a self-tapping screw may tend to cut a new thread with each reinsertion, thereby reducing purchase.

Baumgart and associates showed that insertion torque and pullout strength were comparable for tapped and self-tapping screws. Only if the cutting tip did not protrude through the second cortex did they find a reduction of pullout strength of approximately 10%. Schatzker and associates went on to prove that self-tapping screws inserted at 80% of thread-stripping torque and then removed and reinserted 12 times did not lose any significant holding power.

Locked screws are incorporated in more recent plate designs and may be inserted as unicortical or bicortical screws depending on the type of plate and fracture. These screws produce minimal axial force, if any, and provide fixation by locking the screw head into the plate with a taper thread, perpendicular to the plate. Biomechanically, locking screws function more like a bolt than a screw (see Image 6), and the system acts generally like an internal external fixator (see Image 7). These systems are discussed further in Plates.

Plates are provided in various sizes and shapes for different bones and locations. Dynamic compression plates (DCPs) are available in 3.5 mm and 4.5 mm sizes. The screw holes in a DCP are shaped with an angle of inclination on one side away from the center of the plate. When tightened, the screw head slides down the inclination, causing movement of the bone fragment relative to the plate (see Image 8). As one bone fragment approaches the other at the fracture, compression occurs. The shape of the holes in the plate allow for 25° of inclination in the longitudinal plane and 7° inclination in the transverse plane for screw insertion.

Limited-contact DCPs (LC-DCPs) were designed to limit possible stress shielding and vascular compromise by decreasing plate-to-bone contact by 50% (see Image 11). Theoretically, this leads to improved cortical perfusion with increased preservation of the periosteal vascular network and reduces osteoporosis under the plate. The regular DCP has an area of decreased stiffness located at the plate holes and, with bending, has a tendency to bend at the holes with a segmented pattern, whereas the LC-DCP, with a different geometric design incorporating the holes and plate undersurface, allows for gentle bending distributed throughout the plate (see Image 12). Finally, the LC-DCP is designed with plate hole symmetry, providing the option of dynamic compression from either side of the hole and allowing compression at several levels.

Techniques for the application of both the DCP and LC-DCP are similar (see Image 13). Screws can be inserted in neutral position or a compression position, depending on the desired mechanical result. The DCP uses a green guide to insert a neutral screw, which adds some compression to the fracture owing to the 0.1-mm offset. The gold guide produces a hole 1 mm off-center, away from the fracture, and allows for 1 mm of compression at the fracture site with tightening of the screw. The LC-DCP universal drill guide allows for either neutral or eccentric placement of screws. When creating an eccentric hole to one side or another, the guide is slid to the end of the plate hole without applying pressure and the hole is drilled. By placing pressure against the bone with the drill guide, the spring-loaded mechanism allows for centralization of the hole for neutral screws, particularly if the screw must be inserted at an angle to the plate.

The 3.5 one-third tubular plate is 1 mm thick and allows for limited stability (see Image 14). The thin design allows for easy shaping and is primarily used on the lateral malleolus and distal ulna. The oval holes allow for limited fracture compression with eccentric screw placement.

Improvements have been made for plates used for almost all areas of the body that require placement of a plate near a joint. The refinement of contour, along with screw head modification, reduces hardware prominence and increases fixation options.

The 95°-angled plates are useful in the repair of metaphyseal fractures of the femur (see Image 15), because, if inserted properly, they provide reconstitution of the femoral mechanical axis. Proper insertion requires careful technique so that the blade or screw is inserted with consideration for 3 dimensions (ie, varus/valgus angulation, anterior/posterior position, flexion/extension rotation of plate). The screw barrel devices are considered easier to insert because the flexion-extension of the plate is correctable after insertion of the screw.

Reconstruction plates are thicker than third tubular plates, but they are not quite as thick as DCPs (see Image 16). Designed with deep notches between the holes, they can be contoured in 3 planes to fit complex surfaces (eg, around the pelvis and acetabulum). Reconstruction plates are provided in straight and slightly thicker and stiffer precurved lengths. As with tubular plates, they have oval screw holes, allowing potential for limited compression.

Cable plates incorporate a large fragment plate with cerclage wires to be used with a tensioning device. These are used primarily with femoral fractures surrounding or adjacent to prosthetics (femoral hip or knee implants). Cortical allograft struts are often incorporated for osteoporotic bone.

Diaphyseal plate fixation associated with an anatomical reduction and interfragmentary compression provides absolute stability. Plates are often indicated in articular fractures to neutralize the axial forces on the interfragmentary screws, compressing cancellous bone to facilitate its healing. A fracture anatomically reduced without a gap and fixed with absolute stable fixation will undergo primary healing.

Dead bone at the fracture site is resorbed by osteoclasts of the cutting cones as these cells traverse the fracture site. The osteoclasts are closely followed by ingrowth of blood vessels and mesenchymal cells and osteoblast infiltration. Stress shielding of the bone is rarely caused by the plate relieving axial load to the bone. Plate-induced osteoporosis is caused by disruption of the local vascularity to the bone cortex secondary to an impediment of centrifugal cortical blood flow by the plate. Osteoporosis under a plate should be kept in mind after removal of hardware because the bone also has the mechanical disadvantage of empty screw holes. This vascular-caused cancellization of the cortical bone usually resolves within 2 years of plate application, so it is safe to remove a plate at this time, with the refracture rate being minimal.

Standard plate fixation requires exposure of the fracture site, hematoma evacuation, and reduction of the fracture with possible interfragmentary lag fixation. After a fracture occurs, the periosteal blood supply is dominant, and this network of connective tissue must be preserved to optimize healing. Excessive periosteal stripping and careless soft tissue techniques can impair local blood supply and prolong healing.

Bridge plating is used for comminuted unstable fractures in which anatomic restoration and absolute stability cannot be achieved. Minimal exposure and indirect reduction techniques are used to preserve the blood supply to the fracture fragments for healing, and a plate is attached to the 2 main fragments spanning the area of fracture. The plate is used to provide proper length, axial alignment, and rotation but is obviously limited for any load.

More recent minimally invasive plate techniques use the newer locking plates functioning as an internal fixator with locked screws. Locking plates such as the Less Invasive Surgical Stabilization (LISS) plate (Synthes; West Chester, Pa) are advanced in the submuscular tissue or over the periosteum but do not necessarily contact bone along the length of the plate. This technique limits the disruption of periosteal blood supply that is seen in conventional plating systems. The early Point Contact Fixator (PC-Fix) system (Synthes) and the later LISS system take advantage of unicortical, self-drilling, self-tapping screws with threaded screw heads that lock into the screw hole of the plate.

Once the LISS plate is aligned with the central shaft of the bone, screw placement can be accomplished percutaneously with a radiolucent guide attachment to the plate. Unicortical screws are recommended for use in diaphyseal bone, and longer screws are recommended for use in the metaphyseal area, functioning as a fixed-angle device.

Currently, most manufacturers either offer new locking plate products or they are soon to be released. These devices range from standard straight plates of all sizes, with locking and standard screws, to anatomically specific plates that act as fixed-angle devices. These new plate designs incorporate improved contour with locking screw options for fixation, offering significant advantages over the conventional designs for certain fractures. Proximal and distal humerus, distal radius, distal femoral, and proximal (bicondylar) and distal tibial fractures are examples of injuries that should benefit from this technology, having the improved ability to hold a fracture in its anatomic position and resist applied forces while healing. Conventional plates, which rely on friction forces against the plate from screw fixation and buttressing in metaphyseal and articular fractures, are limited in resisting applied loads versus locking fixation.

In contrast, certain shaft fractures with stable patterns and adequate fixation room already have high union rates with conventional plating (humeral shaft, radius and ulna shaft), and any significant difference between the 2 techniques is difficult to realize. Current recommendations are to use locking screws in situations with limited fixation options, osteoporotic bone, or need for fixed-angle support. For example, a simple lateral plateau fracture that requires buttress fixation and with which the bone quality is reasonable can be adequately treated with a conventional nonlocking lateral plate.

In the near future, most conventional plates will eventually be replaced with plates with locking capability and the decision by the surgeon will consist of deciding which screws are locking or nonlocking, depending on the fracture. As with cannulated screws, note that locking screws can vary in cost ranging from 8- to 15-times the cost of a conventional screw and therefore should only be used when appropriate. This issue is lessened though, when taking into account the need for revision surgery due to malunion; thus, a balance of usage guided by conventional wisdom, common sense, and biomechanical and outcome studies must be obtained.

Plates and other constructs can be used to function as a tension band if an eccentrically loaded bone (eg, the femur) has the device placed on the tension (convex) side of the bone. Using load-strain diagrams, Frederic Pauwels, who first described the tension-band concept, showed that a curved tubular structure placed under an axial load had a tension side and a compression side. With this theory, he described the application of internal fixation on the tension side to convert tensile forces into compressive forces at the fracture site.

With static compression applied by the implant (eg, tensioning of wire, compression with plate), dynamic compression then develops with joint flexion, as with a patella or olecranon fracture, or with load, as with lateral femoral plating (see Image 17). With this technique, the internal fixation device must have the strength to withstand the tensile distraction forces created by muscles during motion, and the bone on the opposite side of the plate must be able to withstand the compressive forces as a medial buttress. Wires and plates are usually quite strong under pure tension forces, but with bending forces added, fatigue can occur rapidly. If bony support is compromised on the cortex opposite from the tension device (eg, from fragmentation, osteoporosis), bending stresses can develop, causing failure of fixation. Wiring techniques commonly include longitudinal K-wires for rotational and axial alignment control in the case of bone fragmentation.

Conversely, fixation on the concave side of the bone occurs in rare situation such as with medial plating of a femur or anterior plating of the humerus. In these situations, fractures have minimal resistance to bending stresses, and gapping can occur on the convex side, resulting in failure of fixation (see Image 18); therefore, attempts should be made to limit potential bending forces to fixation to prevent fixation failure. The tension-band principle can be applied to wires, cables, suture, plates, and external fixators as long as the basic principles are followed.

In the 1930s, Küntscher refined nailing techniques, with the result of IM nails becoming the standard for femoral shaft fixation. Later developments resulted in IM devices being options for proximal and distal metaphyseal/articular fractures and for tibial and humeral fractures. IM nails allow for stable fixation of diaphyseal fractures with early mobilization of joints, early ambulation, and weightbearing of extremities. As metallurgy and designs improved, the indications and techniques for IM devices increased. Specially designed nails for each bone and fracture, IM nails are advantageous over plates and external fixation because the IM location allows for axial alignment and load sharing.

The location and type of fracture determines the device to be used, and devices are named accordingly. IM devices can be described based on dimensions of length, diameter, curvature, locking options, cross-sectional geometry, material, and insertion site options as determined by the bone and fracture being addressed.

A nonlocking cloverleaf Küntscher nail is an example of a centromedullary nail, which is inserted in line with the femoral canal and relies on longitudinal interference with bone-to-nail contact at multiple points to maintain axial and rotational stability of the fracture.

Condylocephalic nails such as Ender pins or Rush rods were a significant device in the early years of fracture fixation. These devices were of smaller diameter and solid. They were inserted in the condyles of the metaphysis, advanced across the fracture, and embedded in the opposite metaphyseal area. These nails were usually inserted in clusters for bending stability but had limitations with rotational and axial forces.

Initial simple IM devices relied on bony realignment with bony contact reestablished along with an interference fit in the medullary canal for stability to axial, rotational, and bending forces. This was enhanced by the cloverleaf designs that added a dynamic lateral compression within the canal for additional stability. As nail designs progressed, interlocking options were added, which improved on the stability and fracture fixation and increasing their indications for use.

Interlocking screws increase the working length of the nail from a simple interference fit, not attainable with nonisthmal shaft fractures or fractures without stable bony contact, to semirigid fixation at the ends of the nail, which is capable of resisting axial and rotational forces. The working length of a nail corresponds to the fracture areas between the sites of fixation and, therefore, can vary from several millimeters with a simple transverse fracture to the entire length of nail between the locking screws in fractures with fragmentation or an unstable pattern. By the 1980s, examples of second-generation interlocking nails included the Grosse-Kempf nail and, later, the Russell-Taylor nail. Currently, all nail manufacturers include basic interlocking screws and other notable features on third-generation nails, such as proximal femoral head/neck screws and dynamic screw slots.

Reconstruction-type nails and gamma-style nails with a reinforced proximal section that allow for fixation into the femoral head and neck region are cephalomedullary nails. These increase the fixation options for proximal femoral fractures. The working length of the nail is increased when the locking screws are located as close to the ends of the nail as is structurally possible, increasing the potential fracture indications.

Tibial nails have also evolved over the years in a similar fashion. With the introduction of locked femoral nails, the same principles of static and dynamic locking were applied to the tibia. By changing nail design and improving the metallurgy, more configurations for locking were possible, thus expanding the indications for tibial nailing to the proximal and distal end segment of extra-articular fractures.

Locking configurations can be static or dynamic. A statically locked nail implies the presence of proximal and distal screws in a nonslotted hole, allowing for control of axial translation and allowing for rotation, with the nail performing more as a load-bearing implant. This application is appropriate for unstable fracture patterns or locations and is certainly a consideration if immediate, full weightbearing is needed, as is sometimes the case in patients with multiple traumatic injuries. As with any fracture reduction, attention to accurate length restoration and rotation is important for avoiding malreduction and leg-length inequalities. Avoidance of fracture distraction is important to minimize the risk of delayed union or nonunion, especially in the humerus and tibia.

Dynamic locking allows the shaft to axially translate several millimeters while rotational control is maintained. This was originally accomplished by leaving the locking screw hole farthest from the fracture empty. This is rarely performed now. Brumback et al demonstrated that dynamic locking leads to malunions, and they recommend static locking for all long bone fractures treated with IM fixation. More recently, nails are constructed with a slotted locking screw hole, allowing placement of the locking screw so that the nail moves along the slot (approximately 5 mm) while the screw controls rotation. With these improved nails, a dynamic option for those fractures with an obviously stable fracture pattern (eg, isthmic location, Winquist fracture pattern II or less) is available to help stimulate healing with axial loading. A statically locked nail may be converted to dynamic lock by removing the static position screws at one end of the nail.

Cross-sectional geometry varies widely with manufacture and design and with fracture indication. Nails may be solid, open-section (slotted), or solid-section cannulated of various shapes, including cylindrical, square, triangular, cloverleaf, and multigrooved or multifluted. Solid nail designs may be necessary for smaller-diameter devices, but they do not allow for insertion over a guide wire and they are difficult to extract if broken. Additionally, recent femoral designs have been replaced with cannulated versions. Bending and torsion strength is altered by changing wall thickness, materials, and, possibly, the number of (adding) channels or slots. A channel along the length of the nail potentially allows for revascularization, but, with the advent of locking screws, the sharp flutes or edges of earlier nail designs were not necessary for rotational control.

Torsion and bending resistance in a cylindrical structure is proportional to the fourth power of its radius. By increasing the radius away from the load axis by a thicker wall or greater diameter, the rigidity increases. Increasing the diameter of an IM nail by 1 mm increases its rigidity by 30-45%, but this would require additional reaming of the canal. Excessive reaming may weaken the diaphyseal bone and increase the possibility of thermal necrosis. For torsion, the rigidity decreases inversely to the working length, and with bending, the stiffness is inversely proportional to the square of the working length; therefore, the shorter the effective working length of the nail fixation and fracture combination, the stiffer the device.

IM implants provide stable fixation, but healing occurs primarily through the formation of periosteal callus. Reaming of the medullary canal increases the working length of an IM implant by elongating the isthmic region with a uniform diameter, thereby increasing the potential implant-to-bone contact. In addition, this allows for a larger-diameter and stronger nail to be inserted than with an unreamed nail, which often allows larger-diameter locking screws, decreasing potential implant failure.

Reaming of the medullary canal damages the medullary vascular system and increases the IM pressure and temperature, with devitalization and necrosis of the diaphyseal cortical bone. In animal studies of blood flow to long bone diaphyseal regions, reaming can cause necrosis of the inner half of the cortex, but this is followed by a strong hyperemic response in the periosteal and muscular blood flow. These changes appear to be reversible over a 12-week period.

Diaphyseal reaming also weakens the bone, and the recommendation is that the cortex should not be reamed to less than half of its original thickness. Additionally, any instrumentation of the medullary canal, including placement of a guide wire and reaming, embolizes marrow contents to various organs, including the pulmonary system. IM pressure can be reduced by the presence of a fracture, slowing the rate of reamer insertion, increasing the speed of the reamer, and allowing the reamer tip to incorporate a small shaft relative to the diameter of the reamer, with deep flutes designed for depressurization of the canal. Although this type of embolization is performed in humans show are undergoing transesophageal echocardiography, its clinical significance is still debated with regard to its affect on pulmonary function in patients with multiple injuries.

Unreamed nailing has been studied as an option to reamed nails, and various studies have demonstrated improved preservation of endosteal blood supply and more rapid revascularization than occurs with reamed techniques. This advantage is limited. Blood flow rapidly improves with reamed fixation, provided the soft tissue envelope is adequate. Most recent clinical studies have revealed improved healing rates for both femoral and tibial fractures (excluding severe open injuries) with reamed nails versus nonreamed nails.

In North America, the standard practice is to insert reamed IM nails in all closed femoral and tibial diaphyseal fractures. The contraindication to this practice is with patients who have been in shock, have pulmonary compromise, elevated serum lactate levels, and abnormal base deficits who also have multiple injuries. Open fractures are also amenable to reamed nails. Grade IIIB open fractures may be a relative contraindication. Humeral nailing still presents problems with union, shoulder stiffness, and neurological injury when inserting locked screws, so it is not as popular as for the other long bones.

Polymers including polylactic and polyglycolic acids and polydioxanone are resorbable suture materials that are currently under continued design and refinement for use as rods or screws that reabsorb with time. These devices offer the theoretical advantage of eventual resorption, eliminating the need for later removal, while allowing stress transfer to the remodeling fracture. Current bioabsorbable implants do not have mechanical properties to match metallic implants; therefore, their indications are limited and their fixation usually requires protection from motion or significant loading. Degradation rates vary, and local inflammatory reactions, such as chondrolysis noted with placement in proximity to joints, have been reported with some implants. These devices are a consideration when fixation of low stress areas is needed and when later removal is anticipated, such as in pediatric patients or in medial malleolar fractures, syndesmotic fixation, or osteochondral fractures in adults.