This invention has to do with implants, devices, methods and apparatus for use in surgery, particularly but not

exclusively surgery for the repair of hip fractures. The invention has particular reference to novel devices, apparatus and procedures for improving any of the efficiency,

effectiveness, safety and acceptability to the patient of the surgical treatment of fractures at the neck of the femur and especially in elderly patients.

BACKGROUND

A variety of methods and devices exist for the surgical treatment of fractures at the neck of the femur (hip

fractures) . A choice among these is made based on the

position of the fracture, particularly as to whether it is inside or outside the joint capsule, the type and extent of damage to the bones, and the general condition of the bone which in turn depends on the age of the patient.

One significant issue is that if the fracture is intra ^¬ capsular the blood supply to the femoral head will be

affected, so that even if the fracture can be suitably reduced and impacted there may be non-union and even subsequent collapse of the structure of the femur head.

Where repair of the fracture is unsuccessful, or unlikely to be successful, the alternative is partial or total hip replacement in which the head and neck of the femur and sometimes also the acetabulum are replaced with prostheses. This is major surgery which carries a high tariff of mortality and morbidity in elderly and unfit patients.

Repair of fractures of the femur neck generally entails driving one or more fixation screws along through the femur neck, through the fracture and into the femur head. The functions of such screws include enabling the fractured surfaces to be impacted together by tensioning the screw to apply compressive force, and holding the parts of the bone in position relative to one another both axially (in the

direction of the neck) and rotationally (preventing rotation of the head around the neck axis) . Known surgical implants for these purposes include the so-called dynamic hip screws, which include a rotatable threaded lag screw to be inserted along through the femur neck from the lateral exterior and a reinforcing reaction structure against which the outer end of the lag screw can act via a screw system to tension the screw and compress the fracture. The reaction structure might be for example an external hip plate to lie against the outer side of the femur, with a hole or barrel formed at a

predetermined angle (corresponding to the angle of the femur neck to the shaft) through which the lag screw is inserted. The hip plate may itself be screwed to the outside of the femur. Or, an intramedullary insert such as an intramedullary nail may be used to provide stronger support and orientation for the screw; this insert fits down into the medullary cavity of the proximal femur to a greater or lesser extent, and has a solid or cannulated head with a through-hole at the

predetermined angle through which the hip screw is inserted.

Intramedullary inserts are used especially when there is damage at the top of the femur shaft. Conversely, sometimes two or three hip screws are used without a discrete reaction structure .

A wide variety of proposals for such hip fracture repair devices and methods exists in the state of the art: see for example US8177786B, US8197484, WO2011/028520, US2012/0191092, WO2009/004603 and US2008/0262498.

While these fracture repair methods can obviate hip replacement which makes them viable treatments especially for very elderly patients, they are liable to subsequent failure which in cases of intra-capsular fracture is caused by

collapse due to non-union at the fracture and/or to loss of blood supply to the femur head.

An important challenge for the surgeon is accurate positioning and alignment of the implants or fixations. The surgeon is generally aided in two respects. Firstly, the implant systems - and especially when these are accompanied by reaction plates or intramedullary devices with holes for engaging and aligning the implant pins or screws - are

normally used in conjunction with external alignment devices in the form of jigs, to ensure exact relative alignment of the components as they are inserted. The devices may also

regulate the degree (distance) of insertion. However these devices assure alignment of the mechanical components strictly only relative to one another, and not in relation to the patient's body. Accordingly surgery is normally carried out under image-assisted conditions using ongoing X-ray imaging (fluoroscopy) by which the surgeon sees real-time images of the surgical site. The preferred mode for hip surgery

provides images for both the AP (anterior-posterior) and lateral directions. Such images enable the surgeon to monitor the insertion of instruments (such as guide wires, drills, trocars etc.) and implants (such as percutaneous hip screws) both as to position within the femur neck (e.g. in relation to one or more other screws which have been or will be inserted) and also as to the extent of insertion so that e.g. the compact bone at the femur head surface is not threatened.

Most usefully the AP and lateral images are shown

simultaneously on adjacent screens. However while this imaging is invaluable for the surgeon the associated

apparatus, especially for lateral imaging, is bulky and significantly obstructs the surgeon so that the benefits of real-time imaging are compromised or not fully available.

A further issue is that lateral imaging requires the patient's unaffected leg to be lifted and held out of the way in a very awkward and uncomfortable position, again promoting prolonged full anaesthesia as the normal operating condition.

Considerable skill is needed to manipulate surgical instruments confidently based on the information from

fluoroscopy images as presently available. Known implant structures may be adapted so that the surgeon can make

adjustments to compensate for some unavoidable inexactitude. For example, intramedullary nails generally have a cylindrical cross-section so that if necessary they can be slightly rotated around their own axis in situ to fit exactly with the associated screw (s). This is an undesirable necessity, since ideally the reaction structure for a hip screw should hold the orientation of the hip screw in all directions, including that corresponding to rotation around the femur shaft, and that would best be achieved by a close form fit. Since that has not been practically achievable, in practice such rotation is ultimately prevented by inserting a locking screw through the wall of the femur shaft and into a hole in the distal part of the nail.

Every complexity and obstruction in the performance of the operation can lead to any of delay, difficulty, a need for larger incisions and a need for greater extent of anaesthesia. All of these factors militate against acceptability and success of the fracture repair operation especially for elderly patients.

THE INVENTION

Taking into account the situation in the present state of the art, we make the following proposals. In particular we have in mind that if a hip fracture repair can be made with minimum delay and minimum anaesthesia while achieving bone repair with adequate reliability, the viability of the

procedure for elderly patients in particular can be markedly improved . Implant Pin

A first aspect of our proposals is a bone fixation implant pin, to be inserted at a fracture site to connect parts of bone across the fracture, and having a shaft portion with an expandable end region which is to be positioned within a bone part typically in a cavity region or region of spongy bone, the implant comprising one or more projection elements movable from an inward position to an extended position projecting at the expandable end region in the radial

direction relative to the axis of the shaft portion.

Desirably there are plural projection elements.

Preferably the (or each) projection element is elongate, and projects into and through the mentioned region in its

direction of elongation, e.g. in a cantilevered fashion as a projecting limb, or in a looped fashion having two ends joined at the shaft portion and the loop projecting. As explained below, the intention is for the projection element (s) to extend into or through a body of injected set compound such as bone cement in the mentioned cavity/spongy bone region, providing a strong anchoring of the implant pin into that region and/or internal reinforcement of the body of set compound by extending through it as tensile reinforcing elements .

Preferably the implant pin has an internal injection cavity such as a longitudinal channel, with at least one injection opening communicating between the cavity and the exterior of the implant pin at or adjacent the end region, for the injection of a settable compound into said bone region via the implant pin. This feature may co-operate with the

provision of the one or more projection elements to form a body of set compound into which the projection elements engage. This is a novel and convenient way to anchor an implant pin into bone across a fracture site, such as through a femur neck, while at the same time strengthening the bone portion - such as the femur head - into which the pin is anchored. While it is correspondingly preferred to be able to inject the settable compound through the same pin or element thereof, an alternative is to inject settable compound

separately either previously through the same bone opening, or through a separate opening.

The shaft may have an internal injection cavity with at least one injection opening communicating between the cavity and the exterior of the pin at the active end thereof for the injection of a settable compound into the spongy bone.

To promote retention in the mass of set compound, the projection elements may be bent non-uniformly, crooked, corrugated, grooved or hooked, in their radially-projecting portions. The projection elements may be operated i.e. moved from the inward position to the extended position via an actuating mechanism operable from an opposite end of the implant pin, such as a rear end or reaction end which in use is at the bone exterior. For example one or more projection elements may be provided initially fixed at the expandable end region of the pin in the inward condition. The actuating mechanism, which may be part of the implant pin as inserted into the bone, may include a drive element which is moveable relative to the shaft portion and operable by an actuating member from the other end of the pin, e.g. by a longitudinal push or pull relative to the pin shaft, to effect a relative movement of the drive element causing the one or more

projection elements to be moved e.g. bent, pivoted or pushed radially outwardly to the extended position. For example the projection elements may be in the form of a set of elements initially generally in line with the shaft portion, but which can be spread out to project radially outwardly (at least, with a radially outward component relative to the shaft axis) . The drive element may engage these elements to push them outwardly, e.g. via a cam surface on the drive element. Or, the projection elements may be carried on the drive element which spreads them by pushing them against a fixed cam or abutment surface of the shaft portion. The skilled person will appreciate that a number of possible mechanisms,

desirably operable by a push or pull mechanical action from a rear end of the implant pin, can be provided to achieve the desired effect. Conveniently a controllable and progressive push or pull action is provided by a threaded engagement between a rotatable inner drive element and a non-rotating counter-element of the drive mechanism at the expandable end of the pin. It can be driven by rotation of the drive element by an appropriate tool. A further possibility is for the projection elements to be held at the inward position against a spring or resilient bending force, and the drive element or actuating movement releases them from this restraint e.g. by moving a catch element or by bringing the projection

element (s) into register with a release opening or window which allows them to spring or open outwardly. We prefer that at least two projection elements, e.g. three to six, are provided. Or, one or more projection elements of complex shape

(of three-dimensional projection path or convoluted shape) may be used.

Desirably the transverse/radial dimension of the implant pin at the position of the extended projecting element (s) - the active end region - is at least twice, at least 2.5 times and more preferably at least 3 times that dimension in the non-extended (insertion) condition. Additionally or

alternatively it may be at least twice, at least 2.5 times and more preferably at least 3 times the largest transverse/radial dimension of the implant pin shaft as a whole (i.e. excepting any external head which does not enter the bone) .

An actuating element for the drive may be provided by or comprise an inner element extending along inside a bore of an outer element of the implant pin in the form of a tubular sleeve or the like. The one or more projection elements may be formed integrally as part of the outer element.

Additionally or alternatively it/they may be provided as part of an inner element, or as a separate element or set of elements restrained relative to the outer element, e.g. inside it, so that in any event when the projection element (s) is/are engaged in the mass of set compound, the implant pin as a whole is anchored.

The implant pin or an element thereof such as an outer element may be formed with a cutting tip or annular cutting edge so as to be self-drilling . Self-drilling pins/elements have the advantage of reducing the number of steps, provided that the situation allows for reliable insertion at the correct position and alignment. However, the pin may equally well be inserted into a pre-drilled hole which may be formed by known techniques, e.g. with the benefit of a guide wire.

The implant pin may or may not carry an external thread. In some embodiments it does not carry an external thread and relies solely on the projection elements for anchorage against withdrawal. In other embodiments it may have an external thread such as a self-tapping or lag screw-type thread.

The designs of implant pins proposed herein are

independently novel. They may have uses for fixation which do not require the expandable formations to be embedded in a mass of set compound such as bone cement.

Use of the Implant

A second aspect of our proposals is a bone fixation procedure for use at a fracture site, particularly but not exclusively at a hip fracture, to connect parts of bone across the fracture. The method comprises inserting a bone fixation implant pin as described above, with the expandable end region thereof positioned in a bone part, typically in a cavity region or a region of spongy bone therein, injecting a

settable compound into the bone part to form a mass

surrounding the end region of the implant pin, moving the one or more projection elements of the implant to the extended position (before, after or during the injection of the

settable compound) to be embedded in the body of settable compound, and setting or curing the settable compound to form a solid body of set compound inside the bone part with the end region and one or more projection elements embedded in the solid body to anchor the implant pin.

Preferably the projection elements of the implant are actuated to the extended position before injecting the

settable compound. Preferably the settable compound is injected through the implant pin, e.g. via injection openings at or adjacent the end region of the implant pin communicating with an internal injection cavity of the implant pin.

The settable compound is desirably a curable polymer compound. A variety of these is known to the skilled person, and it may be in the nature of a bone cement; typically a compound based on acrylic polymer, such as methyl methacrylate monomer to form PMMA, may be used. The compound may contain solid particulate filler, such as cured polymer beads or a powder, in the curable monomer mix. The compound composition may be selected and adapted to optimise its mechanical

properties (when set) for the characteristic purpose of forming a substantial retaining and support body inside bone. Relevant properties include any of toughness, crack- resistance, tensile strength and compressive strength.

Biocompatible or inert additives or filler materials for these purposes are known, such as hydroxyapatite and carbon fibres or carbon nanoparticles such as nanotubes.

The method is particularly preferred for anchoring into the femoral head, and the mass of set compound may fill or substantially fill an interior cavity or spongy bone region of the femoral head.

The injected compound may adequately permeate the open- cell spongy structure, especially if this is deteriorated such as in an elderly or osteoporotic patient. Optionally the method includes a preliminary step of removing or clearing bone material around the expandable end of the implant pin - or, more usually, the intended location thereof - by a

mechanical and/or lavage method, to promote the formation of a continuous solid body of the set compound with the projection elements of the implant pin embedded in it. This method has the significant advantage that a continuous body of the set compound can substantially or entirely occupy or fill the bone area, which might otherwise be fragile or of low quality, and give substantial and lasting mechanical support to the compact bone joint surface of the femoral head. A second advantage is that it provides a widely-distributed mechanical engagement for the implant pin, giving good stability for the repair and a low probability of collapse, because the set mass of

compound can support the repair in both directions. Finally, the preferred form of projections on the implant pin, i.e. an elongate form exposed to be embedded in and surrounded by the compound mass, or transfixing it, can act as tensile

reinforcements for the compound mass further strengthening the repair .

The use of some bone cement around a bone screw is of course a known practice. However the formation of an enlarged body or mass of set compound, penetrated by one or more projecting elongate retaining and reinforcing elements of an implant pin, and which may substantially occupy the head of a bone to constitute a primary mechanical element of the repair, is a novel proposal.

One preferred option for clearing a void around the implant pin head location, e.g. within a femoral head, is by insertion of an inflatable balloon element in a collapsed condition, followed by inflating the balloon element to expand into the intended void location, clearing fragile bone

material away. Depending on the structure of the implant pin and the sequence of operations intended, this balloon element may then be deflated and retracted, leaving the cleared void ready for occupation by the injected settable compound, or it may be left in situ and itself filled with the settable compound. Principles and techniques for use of inflatable balloons to clear or enlarge a void in cancellous bone are known in themselves (e.g. in the treatment of collapsed vertebrae) and saline can be used for the inflation,

optionally via a catheter. Techniques and means are known for limiting or controlling the degree or shape of expansion of the balloon to protect the surrounding bone structure and for enabling the use of imaging to follow the operation. Any of these means may be used in line with the skilled person' s knowledge to adapt them to the present purpose. Or, an insertable and withdrawable mechanical crusher device may be used to clear a local void adjacent the bore.

In one preferred form of the procedure the implant pin or outer sleeve element thereof is inserted - optionally in a self-drilling and/or self-tapping mode - into the bone, and has a central cavity or channel providing access to the bone interior from the exterior working site via one or more holes at the inserted end. An optional stage of removing bone material from around the inserted end of the implant pin may be carried out through the implant pin itself e.g. by a pressure or lavage system, or by insertion of an inflatable balloon as described above. Or, an insertable and withdrawable mechanical crusher device may be used. The implant pin is actuated to extend the projection element structure, and the settable compound is injected, preferably through the implant pin. Usually the projection element structure is extended before injecting the settable compound. Where the initial stage is done via an outer element of the implant pin, an inner element comprising means for injection and/or the projection element structure may be inserted as a separate step. Since plural fundamental steps of the fracture repair can be done by or through the same implant element, significant time can be saved.

A bore in which the implant pin is inserted, either predrilled or by self-tapping insertion of the pin, may be positioned in relation to the femur such that the side of the pin engages the calcar. Such positioning gives valuable load- bearing support to the pin, and is facilitated by precise imaging during the implant procedure.

A further option is the use of an occluding member, device or formation around the implant pin to inhibit or block undesired flow of settable compound out into the fracture. This may be a component in the form of a tube or sleeve to provide a local enlargement of implant cross-sectional area or a blocking of any gap at the fracture line, or at least between the fracture line and he intended site of forming the set mass of compound. It may be a discrete component. It may be inserted before the pin, or on or as part of the pin. It may be expandable out into contact around the bore for the implant. It may be expanded by the expanding action of the projection elements at the expandable end of the pin.

A further option is that an internal channel and

injection holes of the implant pin can be used to introduce local anaesthetic during the procedure.

It will be understood that while an implant pin and procedure have been described primarily with reference to repair of a hip fracture because this is of great importance, the device and method proposals are applicable in other procedures, such as in fractures of the humerus, and the anchoring mode of the implant into a mass of set compound around the end of the pin may be applicable also in other procedures. Bone fixation methods in which the implant pin head is expanded inside a bone without embedding in a mass of settable compound are also contemplated herein. As with known implant pins and fracture repair screws, the present implant pins may be used readily in conjunction with reaction

structures such as plates and inserts, e.g. hip plates and intramedullary inserts or nails. Further proposals for forms of reaction insert are made below.

Imaging System/Imaging Method

A further proposal herein relates to an imaging system and imaging method intended to improve the availability of real-time imaging in more than one direction/plane with relatively less obstruction of the surgeon and/or less awkward positioning for the patient. The X-ray imaging used during surgery requires an X-ray generator and corresponding detector to be positioned on opposite sides of the patient, with the surgical site between on the imaging axis. The apparatus is large and heavy; usually the generator and detector are supported at opposite ends of a C-shaped arm which can be manoeuvred under power or manually to image along any of a range of selectable imaging axis directions. For example, for AP imaging of a supine patient such as for a hip fracture repair, the ends of the C-shaped arm are respectively above and below the surgical table. For lateral imaging they need to be to either side of the surgical table, generally in the coronal plane, and this is a significant obstruction for the surgeon especially if both AP and lateral imaging are done together, so that the benefits of real-time two-axis imaging are offset by the difficulty of surgeon access.

In general terms, our proposal is to image a surgical site along at least first and second imaging axes, the first and second axes being non-parallel, to obtain first and second primary image data. Because the first and second axes are non-parallel, and each set of primary image data represents two-dimensional image information, the first and second primary image data together constitute three-dimensional image information. In the known AP and lateral imaging for example this is presented simply by showing corresponding first and second displays of the first and second image data for the surgeon to use. Our proposal is to process the first and second primary image data in combination to produce at least one

reconstructed secondary image data output for a display, corresponding to an imaging axis different from the first and second axes. Desirably the first and second primary image data are processed in combination to produce at least two image data outputs for at least two displays, at least one and optionally at least two of these being reconstructed secondary image data output (s) for reconstructed display (s)

corresponding to a notional imaging axis/axes different from the first and/or the second imaging axes.

Optionally, third primary image data may be obtained additionally for a third imaging axis non-parallel to the first and second axes. The third primary image data may be processed with the first and/or with the second primary image data for producing reconstructed secondary image data for one or two or more secondary displays corresponding to a notional imaging axis/axes different from any of the first to third imaging axes and preferably different from all of them. A particular advantage of obtaining data along a third imaging axis is that it may be optimised for a different aspect of the reaction site, e.g. it may avoid some obstruction which the second axis does not. It will be understood that the primary image data from any one or more of the first, second and optional third imaging axes may also be presented as a

corresponding primary display, i.e. a display corresponding to the actual imaging axis such as an AP axis.

Preferably the first and second axes differ by at least 25° or by at least 30°, more preferably at least 35°. If there is a third axis preferably it is angled to at least one of and preferably both of the first and second axes by at least 20°, preferably at least 30°, more usually at least 35°.

The first and second axes are preferably non-orthogonal. Preferably they are at least 10°, at least 15°, at least 20° or at least 25° or at least 30° away from orthogonal. However in some cases they may be orthogonal, and valuable for

obtaining reconstructed image data for a useful notional (non- orthogonal) imaging axis which for some reason is not

practically or conveniently directly imaged.

The benefit of these imaging proposals is that they enable display of images corresponding to viewing directions which need not correspond to the actual direct imaging axis of the imaging apparatus. Provided that two non-parallel axes are available, three-dimensional data are effectively

available which can be used to reconstruct or reconstitute an image corresponding to a different axis of view. Importantly, the disposition of the primary imaging axes can then be chosen to reduce or avoid obstructing the surgical site, although a view displayed may be one which previously would have required imaging along an axis that would obstruct the surgical site. Many operations require the surgeon to act primarily along one main sector of activity. A first primary imaging axis

perpendicular to this should not obstruct the surgeon, but a second axis orthogonal to the first - as would be the

conventional mode - will either obstruct the surgeon or be prevented by the layout of the patient's body. The present proposals enable the second axis to be chosen obliquely, and optionally supplemented with a third axis also chosen

obliquely, reducing obstruction. For example in the context of repair of a hip fracture, the patient lies on their back (supine) on the operating table with the affected hip

generally towards one side of the table. The surgical

activities will mostly be directed in the coronal plane, so arranging a first vertical imaging axis (X-ray generator and detector vertically aligned above and below the surgical table) for anterior-posterior (AP) imaging is convenient and indeed conventional. Direct lateral (orthogonal) imaging of the affected hip joint is valuable for the surgeon but

requires the unaffected leg to be lifted steeply out of the way, which is awkward and, as mentioned, if the dual imaging is to be continuous the lateral X-ray apparatus obstructs the surgeon. In our proposal a second imaging axis is arranged obliquely (non-orthogonally, non-horizontally) with one of the generator and detector above the table level and the other below. By lowering part of the apparatus and imaging at an angle not directly across the table, obstruction is reduced or avoided. The second axis may be oriented to optimise data collection for the surgical site in question e.g.

perpendicular to the femoral neck or femoral shaft, while also being non-horizontal and non-vertical (oblique) . A third axis may be added if desired e.g. perpendicular to the other of the femoral neck and femoral shaft, while also being non- horizontal and non-vertical (oblique) , providing a comprehensive 3-dimensional data set.

Image processing software can be used to manipulate the available data to provide a reconstituted image display as if viewed view along any desired direction relative to the imaged cite. The principles and implementation of such image

processing software for medical imaging data such as

radiographic data are known as such to the skilled person in image processing, it should be said. For example the

reconstruction of image data taken from various imaging angles to form 3-dimensional image data is known for pre-operative diagnosis - see e.g. EP1782734/US7712961, US2015/0049856 and US4138721. For this however the imaging device (mobile C-arm) is scanned around the body to collect the data. Such movement is not usually appropriate for live surgical assistance or guidance. Our proposal prefers concurrent imaging along plural obliquely-angled axes which preferably are not moved or scanned during surgery.

A further and particularly practical option is to

reconstruct three-dimensional image data using "live" data for an image axis imaged during surgery, on an ongoing basis, in combination with previously-stored data for another imaging axis. Such stored data for the other axis might be data for the same patient, taken before the operation starts. Or, they might be standard or library image data (real or artificial) selected to match the patient's anatomy. In the context of hip surgery, for example, the AP imaging may be done in real time, while lateral image data are contributed from stored data as described, enabling data reconstruction to create 3D data useful for purposes described herein such as display, instrument navigation and the like.

Thus, the surgeon may be shown for example a pair of displays with an AP view and an orthogonal lateral view, although there is no direct lateral imaging. The direct imaging along the two or three axes may be continuous or concurrent so that real-time or nearly-real-time images are practically available. Other views may be preferred according to the surgeon' s preference or the circumstances of the surgery. As is known, the imaging system may detect and/or display the positions of instruments (such as scalpels, guide wires, drills and prostheses), and/or their motion paths, either detected directly by the imaging or superimposed on the displayed image; again methods and means are known to the skilled person for this such as sensors or emitters on the instruments for motion and/or position, coupled with

complementary components of the imaging system. The system may be programmed to provide a dynamic display with variable notional imaging/viewing axis for the surgeon to view the surgical site e.g. as seen along the motion path of an

instrument. Or, the system may use the 3D data for guiding instruments in automated or partly-automated procedures;

compare EP1815814.

Our proposals extend to imaging methods as described, and also to imaging systems comprising plural imaging devices (especially X-ray devices, but other imaging modes such as CT may be treated similarly) with respective imaging axes, and a processor programmed with image processing software for carrying out the image processing described, with outputs for one or more displays including the one or more reconstructed displays .

By providing real-time 3D imaging with less obstruction from the imaging system, a surgeon' s accuracy and operating speed can be improved with less need for complex or awkward patient positioning. Surgical instruments, implants and prosthesis can be more accurately positioned. Surgical incisions may be smaller, with less patient trauma. Surgical methods comprising the use of the imaging methods are a further aspect of our proposals.

Reaction structure: insert

A further proposal here in relation to fracture repair, especially for femoral or humeral fractures, relates to a reaction structure adapted to be inserted into the bone along its length to provide engagement for one or more bone screws or other implant fixations, such as an implant pin as

disclosed herein. These insert structures such as

intramedullary nails are widely known and used, and as mentioned previously they are generally made round and smooth in cross-sectional shape so that the surgeon can turn them in the bone cavity for adjustment of the positions of the implant elements. Adjustability is necessary because (as explained earlier) with conventional imaging a surgeon cannot expect to align the insert initially at the precise orientation needed for accurate insertion and locking of the fixation screw, implant pin and the like.

Our proposal is an insert structure to serve as a

reaction structure for a bone fixation device or system, such as with an implant pin or bone screw, shaped and dimensioned to be inserted into the end of a long bone, such as the femur or humerus, and having a guide formation, typically one or more obliquely-angled through-holes, for engagement with a bone fixation device. The insert has a length axis which in use is generally aligned with the length axis of the bone. According to our proposals the cross-sectional shape of the insert transverse to the length axis is oblong, at least at a portion of the insert which in use is seated in a metaphysis portion of the bone, to restrict rotation thereof in the bone.

This oblong cross-sectional shape desirably features a long cross-sectional dimension which is at least 1.3 times, more preferably at least 1.5 times or at least 2 times, the cross- sectional dimension in the transverse direction at the same axial position. The outer surface of the insert at the oblong region may feature flat or relatively flat surfaces on the longitudinal sides (in the cross-sectional sense) thereof. The other surfaces (end surfaces) may be more rounded, or may also be flat but will be general shorter transversely.

The metaphysis of a long bone such as the humerus or femur generally features an enlargement or divergence of transverse cross-section so that by fitting a non-round or oblong-section insert head into a fitting recess of this region, rotation around the length axis can be substantially or entirely prevented. This gives a corresponding more positive location for a bone fixation element mounted through the insert. In relation to femoral inserts, we particularly prefer that a head portion of the insert be at last partially oblong in the coronal direction, i.e. extended towards the neck and head of the femur, so that the oblong region may extend towards or partly into the neck region. In preferred constructions the insert has a region below an insert head which is reduced in cross-sectional size, i.e. downwardly generally convergent in form, to fit into the regions of the bone towards or into the shaft (diaphysis) . A desirable feature is that where there is a medially-directed oblong formation of the insert head cross-section, a medially- directed overhanging (medially and distally-directed) surface portion of this may be oriented and positioned to contact against the calcar region of the bone (on the inside of the neck angle) , which is strong and gives good mechanical

support. Preferably such an oblique surface portion to abut the calcar is angled at between 15° and 50° relative to the opposite face of the insert, i.e. the lateral surface for a femoral insert, or relative to the axial direction of the insert corresponding to the bone axis. A medial surface portion to the proximal side of said oblique surface portion desirably makes a smaller angle with, or is parallel to, that opposite surface or axis. The oblique surface portion may be concave e.g. to fit more closely to the calcar. That is, a part towards the proximal end of the insert may be more steeply inclined from the axis than a part distal thereto.

Additionally or alternatively a lateral face of the oblong insert may have a distally-inclined surface portion to conform to or lie against the diverging portion of the lateral cortex at the metaphysis. The lateral cortex at this position may optionally be shaped in operation to facilitate this.

An insert of this type requires accurate fitting in order to be sure that the guide opening for the bone fixation is precisely aligned with the intended direction of the bone fixation. However, this alignment is achievable, and becomes more reliably obtainable especially by means of the superior imaging methods/systems described herein. When strong

rotational orientation (relative to the bone axis) can be achieved at the head of the insert - which was not normally possible in standard insert devices such as intramedullary nails - it becomes less necessary for the lower part of the insert to extend far into the diaphysis in the medullary space, because less frictional surface is needed and an anti- rotation screw may not need to be driven into the insert from outside the bone. Accordingly the insert can be made smaller and less invasive. Also, the thickness in the transverse direction (anterior-posterior, for a coronally-oblong insert) can be made smaller than before.

The oblong cross-section may continue up to the head of the insert. One or more guide openings for fixation devices

(screws or pins) may pass through the oblong portion. This portion may also comprise a mechanism, or elements thereof, for engaging with a fixation (screw or pin) passing through to drive it along its length relative to the insert for

compressing a fracture, typically. Such mechanism may

comprise a supplementary screw member, ratchet device or the like. Optionally, a cross screw may be used in order to enhance torsional stability.

A surgical method or method of bone fixation comprising the positioning of a reaction insert of the kind described in a bone, particularly in the end of a long bone such as the proximal femur or humerus, is an aspect of our proposals. The method may include cutting a shaped recess inside the bone to fit the insert. The method may include disposing an above- mentioned oblong-section insert with the oblong-section portion thereof in the metaphysis region (optionally also the epiphysis region) of the bone. The long axis of the oblong- section portion may be aligned with the coronal plane. At least one bone fixation screw or pin may be passed through the insert and engaged with or in bone beyond the insert.

Compression may be applied to the bone by tensioning the screw relative to the insert.

Bone Crushing Device

A further proposal herein is a bone crusher device which can be used for enlarging a local void around a drilled bore, such as in the head of the femur. A device of this kind is known for use with fixation systems as disclosed in WO2009/004603. It has a handle and an insertion rod (in fact a tube) for insertion into the drilled bore, with a crushing tip mechanism which, on operation of a lever adjacent the handle which slides an operating rod along inside the tube, pushes opposed crushing members at the tip radially outwardly to crush cancellous bone to either side of the drilled bore. By rotating and repeatedly operating the device, a local cavity can be formed all around the bore.

We have noted that in the context of proposals herein, it may be advantageous to form a local cavity which extends eccentrically relative to the drilled bore. For example, it may be desired to localise a set mass of compound

eccentrically to one side of the implant pin. To that end we propose a device of the kind described in which the crushing tip mechanism has a dynamic crusher element or crusher shoe which moves radially out relative to the tip of the insertion rod, but wherein the tip portion oppositely directed relative to the dynamic crusher shoe is a static or fixed tip portion so that the crushing effect is one-sided. The static tip portion is desirably radially in register with the rest of the rod on that side, e.g. as a continuation thereof, so that it is supported against the wall of the drilled bore on that side over an extended length and does not exert localised crushing pressure in reaction to the crushing action on the other side of the tip.

BRIEF DESCRIPTION OF THE DRAWINGS

The invention is now explained in more detail by

reference to examples and embodiments shown in the

accompanying drawing figures, in which:

Figs. 1(a) and 1(b) are schematic views showing different fracture positions at the neck of a femur;

Figs. 2(a) and (b) are lateral and frontal views showing imaging directions and regions for a hip fracture;

Fig. 3 is a schematic plan view showing patient position on an operating table for lateral imaging during conventional hip fracture repair;

Figs. 4(a) to (d) show stages of clearing a void in a femoral head using an inflatable balloon;

Fig. 5(a) and 5(b) are exploded views showing the

components of a bone fixation system including an implant pin and two different forms of reaction structure in a first embodiment ;

Fig. 6 shows the components of the first embodiment in preliminary position for securing a hip fracture;

Fig. 7 is a detailed cross-sectional view showing part of an actuating mechanism of the implant pin;

Fig. 8 shows the implant pin with the head expanded into a void of the femoral head filled with bone cement;

Fig. 9(a) shows the outline of a novel form of reaction insert fitted in the top of the femur, while Fig. 9(b) shows a screw mechanism for compressing a fracture with it;

Fig. 10 is an exploded view of a second embodiment of bone fixation system including a reaction structure and an implant pin; Fig. 11 shows the components of the bone fixation system assembled together as they would be in situ;

Figs. 12(a) to 12(f) show stages of a procedure for repairing an intra-capsular hip fracture using the fixation system of the second embodiment;

Figs. 13(a) to 13(f) show stages of a procedure for repairing a hip fracture using a third embodiment of implant pin;

Figs.14 (a) to 14(c) show stages of a procedure for repairing an extracapsular hip fracture using an implant pin similar to the second embodiment, and

Fig. 15(a) is an isometric view of a fourth embodiment of implant pin, Fig. 15(b) being a sectional view thereof, and Figs. 15(c) to (e) being respectively a sectional view of its outer barrel, a perspective view of its expansion driving body and a perspective view of its inner element;

Figs. 16(a) to (j) show schematically stages of a

procedure for repairing an intra-capsular hip fracture using the implant pin of the fourth embodiment;

Fig. 17 is a perspective view of a device for locally crushing bone adjacent to selected regions of a drilled bore;

Fig. 18 is a cross-sectional view of the Fig. 17 device;

Figs. 19(a) and (b) are enlarged views of the crusher tip of the Fig. 17 device showing the mechanism, Fig. 19(a) being an external view and Fig. 19(b) being a sectional view, and

Figs. 20 and 21 are respectively a plan view and an end schematic views of an operating table showing imaging axes.

DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS

Fig. 1 shows general features of the proximal femur and some typical fracture types. The proximal femur 900 has a head 901 connected to a shaft 903 by a generally cylindrical neck 902. The angle of the neck 902 to the shaft (diaphysis) 903 varies among individuals, but typically is from 130°-145° usually 135°. The angle portion at the epiphysis is enlarged laterally by the greater trochanter 904, while the lesser trochanter 905 projects inwardly and backwards. Fig. 1(a) shows the line of a fracture 911 just below the head 901 - a sub-capital fracture - and this is an example of an intra- capsular fracture because it lies within the joint capsule

(not shown) . The epiphysis region is otherwise largely intact, so that if the head can be pulled against it and prevented from rotating, the fracture can be reduced. Fig. 1(b) shows an extra-capsular fracture 912, in this case an intertrochanteric fracture, a feature here being that the blood supply to the femoral head is intact, so that if the fixation device can engage with the shaft 903 of the femur to assure proper orientation, the fracture will unite. This is all standard knowledge for the skilled person.

Figs. 2(a) and 2(b) show schematically the need for imaging of a patient P for operating on the affected hip AH with a hip fracture. The patient P lies supine. Anterior- posterior (AP) X-ray imaging (fluoroscopy) is easily done on a vertical axis 161 between generator 160 and a detector (image intensifier or flat detector 162) beneath the operating table, which is generally transparent to X-rays. To operate on the hip the surgeon usually needs lateral as well as AP imaging and this presents problems, because the unaffected hip lies in line with the affected one. In practice, as shown in Fig. 3 the patient's unaffected leg UL is lifted and then held sideways out of the way strapped on a leg support 151 - the affected leg AL being immobilised by a foot support 152. The lateral X-ray arrangement (generator 171 and detector 172) is arranged to image on the horizontal axis generally

perpendicular to the femur neck. Typically the generator and detector of an X-ray imaging system are carried at the

opposite ends of a mobile C-beam which can be pivoted in three dimensions to the desired orientation. If C-arms for both vertical and lateral imaging are positioned as shown, the surgeon's access to the affected hip AH is seriously hindered. In particular, the lateral imaging system obstructs the site.

Later, we describe proposals for improving this.

With reference to Fig. 4, the head of the femur has an outer covering of compact cortical bone 907 while the interior is filled with spongy or cancellous bone 906 of open

structure. As explained above, our proposals include filling a void in this spongy or cancellous region 906 with a mass of set compound for engagement by an implant pin. Fig. 4 shows one preferred method for preparing the spongy bone region to maximise the efficacy of the filling with a set compound. In this method the cancellous bone - which particularly in an elderly patient may be weak and degraded - is cleared out of the way making a fully open void, to create in turn a large, solid (generally continuous) and integrated body of set compound which will form a good anchoring point and also support the compact bone of the joint from within. To achieve this, as shown in Fig. 4(a), a hole 190 is drilled medially from just below the greater trochanter 904 along through the centre of the femur neck, past the intra-capsular fracture 911 and into the head of the femur, finishing at a termination point 191 just short of the cortical bone. Methods for drilling such a hole to a predetermined depth are well- established .

Fig. 4(b) shows an elongate inflatable balloon 180 inserted into the drilled hole 190. Inflatable balloons of this type are known in general, such as for use in the

procedure called kyphoplasty in spinal surgery, where they are inflated with saline under pressure, and similar materials and methods may be applied here. As the balloon inflates (Fig. 4 (c) ) it compacts the soft spongy bone out against the

cortical bone 907. The balloon 180 can then be deflated as shown in Fig. 4(b) and withdrawn, leaving essentially an open void 908 inside the head of the femur.

In the example of Fig. 4 the bone away from the fracture is generally intact, and the balloon can be inflated all along the drilled hole 190. If the bone is more damaged, so that the void should be cleared only inside the head region of the femur without outward pressure elsewhere, a shorter balloon can be inserted at the end of a rigid catheter tube and inflated only inside the femur head; again this can be done with saline and is a known procedure in kyphoplasty: see e.g. WO2013/096197 and the earlier documents referred to on page 1 thereof .

Next, a first embodiment of bone fixation system is described with reference to Fig. 5(a) . It comprises an implant pin generally designated 1 and a reaction structure or epiphyseal/metaphyseal insert 4. The insert 4 is a generally solid downwardly-tapered piece for insertion into the top of the proximal femur at the front of the greater trochanter: see Fig. 6. It has an angled guide opening 42 extending through it, which receives and fixes the angle of the implant pin 1.

A feature of the insert 4 is that it is relatively axially short, not extending far down into the medullary cavity of the femur. Rather, it is externally shaped to engage more rigidly inside the metaphysis, by means of precise positioning in an accurately cut recess in the metaphysis. The insert 4 has a top (head) end 411 with a threaded bore 45 for manipulation, a bottom end 412 of smaller cross-section, and a main body 413 tapering continuously from the top end to the bottom end with a cross-sectional shape which is generally oblong in the anterior/posterior direction, presenting flat lateral and medial faces 43 with more sharply rounded end surfaces 44. The angled guide opening 42 accurately matches the intended angle between the insert and pin in situ; this is carefully determined in advance with reference to the

patient's bone structure and a range of inserts with different guide opening angles may be available for this.

Fig. 5(b) shows an alternative and generally preferred form of epiphyseal/metaphyseal reaction insert 204; the disposition of the guide opening 242 is similar but the cross- section of the block is generally oblong in the medial/lateral direction, and the medial face has a lower face portion 2431 which is convergent or downwardly-directed, that is, angled laterally relative to an upper portion 2432 of the medial face. The effect of this is discussed later with reference to Fig. 9.

In each of Figs. 5(a) and 5(b) the implant pin 1 consists of an outer element or barrel 2, an inner element or drive stem 3 - tubular like the outer element - and an expansion driver body 5. The outer barrel 2 is generally tubular, having an enlarged retainer head 21, a main shaft portion 26 and a tip region 24. The outer shaft 26 is generally smooth on the outside, i.e. without an external thread although a thread may be provided in some variants. The tip region 24 is cut and shaped to define four straight axial limbs 25, which are continuations of the outer shaft tube wall, windows or spaces 23 defined between these, and four expansion arms 27 occupying these spaces, each having an inward bend 271, an outward bend 272 (see Fig. 7) and a free end 273, with the portion 274 adjacent the free end being outwardly inclined.

The inner element 3 fits along inside the outer barrel 2 and is about the same length, also having an enlarged

retaining head 31, a shaft portion 36 and a tip 34. The part near to the tip has a thread 35. An internal bore 32 extends from the head 31 to just below the tip 34, where it is closed off forwardly but opens out through the element wall through a set of injection holes 33.

The driver body 5 screws onto the tip 34 of the inner element 3 and fits closely inside the end limbs 25 of the outer barrel 2 - see Fig. 6 - so that the limbs stop it from rotating. It has a side surface 51, a threaded bore 52 and an inclined cam surface 53 directed back towards the head. The action of the mechanism can be understood from Figs. 6 to 8. In the initial condition of Fig. 6, when the implant pin 1 is inserted into the femur neck so that its tip enters the femur head, the whole pin in front of the retainer head 21 has the diameter of the outer shaft 26. The expansion driver body 5 is threaded onto the extreme end of the inner element 3, and its cam surface 53 just touches the free ends 273 of the expansion arms 27 (Fig. 7) . When the inner element is turned, the threaded engagement draws the driver body 5 back along inside the limbs 25 of the outer barrel 2. Its leading cam surface 53 pushes the expansion arms 27 out, sliding against their inclined portions 274, until they reach the projecting position shown in Fig. 8.

The overall procedure is now described, because this is not completely shown in Figs. 6 to 8. First, a hole is drilled and an enlarged head void 908 formed, e.g. as

described with reference to Fig. 4. A recess 195 shaped to receive the insert 4 is cut out into the top of the proximal femur at this time, intersecting with the drilled hole 190. This may be done before or after forming the void 908. By precise imaging (explained later) the insert 4 and its recess 195 are positioned precisely so that the implant pin 1 will be at the correct angle in the femur neck. An annular abutment spacer 49 (see Fig. 5(a)) is selected; this has an angle face 491 corresponding to the angle of the insert guide opening 42 with its side flat 43, and a perpendicular abutment face 492 against which the implant pin's retainer head 21 will rest.

The pin is passed through the angular abutment spacer 49 before being passed through the insert 4 and through the femur neck. The inner element 3 is then rotated by a suitable tool to move the expansion arms 27 to the projecting position, as shown in Fig. 8. Then, a settable compound such as a bone cement is injected through the central bore 32 of the inner element 3 and out through the injection holes 33 to fill the void 908 inside the femur head. This bone cement then sets solid around the expanded tip of the implant pin as a body 909 which fills the femur head, forming an anchorage which

reliably and permanently locates the implant pin within the femur head to provide a solid reaction for reducing and impacting the fracture in the neck.

It will be understood that, while this implant pin is shown used in combination with an epiphysis/metaphysis insert 4, depending on the nature of the fracture and the extent of bone damage, it may be used with a different kind of reaction structure, such as an intramedullary nail, or with an external hip plate (which may correspond to an extended form of the spacer 49), or without any reaction structure.

For rotational stability, the engagement of the

expandable arms 27 in the solid mass 909 of injected bone cement provides firm rotational engagement, which may suffice if the head end of the pin is also adapted to fix rotationally into the bone, or into the reaction insert where used.

Additionally or alternatively, the present implant pin may be supplemented with one or more additional fixation elements such as another bone screw, pin or wire inserted next to it to inhibit rotation of the femur head at the fracture. A skilled person will understand the options here.

Fig. 9 illustrates further our new proposal for a

reaction insert, such as an metaphyseal or intramedullary insert, useful in conjunction with the present implant pin proposals, but independently useful in conjunction with implant pins and implant screws of other types. See also Fig. 5(b) . With reference to Fig. 9(a), the outer compact bone

(cortex) of the femur diverges at the metaphysis. At the medial position the cortex is thick and strong, and called the calcar 915. The outward divergence of the lateral cortex is also indicated at 916. The insert 304 has a proximal end (head) 345, a distal end 312, medial face 343 and lateral face

347. It is oblong in the coronal plane: its shape may be as shown partly in Fig. 5(b) with rounded edges onto the side faces. Fig. 9 is showing a schematic outline only. An opening 920 is cut in the end of the bone for insertion. The proximal end of the insert 304 then lies roughly in register with the top of the bone (epiphysis) while the distal end 312 extends down slightly into the medullary cavity. The lateral face 347 of the insert is generally straight in the axial direction of the bone, and lies against the lateral cortex at and below the metaphysis. The medial face 343 has an upper portion 3432 generally parallel to the bone axis, facing onto the inside of the femur neck, and a lower (distal) portion 3431 which is inclined at about 20° to the axis, i.e. distally convergent, so as to lie against the calcar 915. The insert 304 makes a positive mechanical shape engagement with the interior of the bone, being held against rotation and also strongly supported by the calcar. An option is to make the calcar-contacting portion concave, to enhance the contact area and support.

Another option is to provide a corresponding concavity or distally-inclined surface portion on the lateral side, to engage against the lateral cortex 916.

Fig. 9(b) shows how in practice the insert 304 will have an angled through-hole 342, to take a bone screw 1001 for a fracture of the femur neck, for example. A mechanism is also illustrated for pulling on the bone screw 1001 to compress the fracture, e.g. by a second screw 1002 shown in a supplementary screw opening 3421. The skilled person is aware of a number of possible mechanisms for this. Holes 921 are drilled through the lateral cortex for inserting the screws 1001,1002. Since the lateral cortex is often cut here, a possibility is to shape it for enhanced engagement with a correspondingly shaped lateral surface of the insert 304.

The descriptions here of the shape of the insert's medial and lateral surfaces refer primarily to their shape in the cross-section in the coronal plane as in Fig. 9. Outside that plane, in the interior and posterior directions, the insert

304 may be rounded off as seen in Fig. 5(b) so that e.g. a calcar-contacting surface may be concave in the coronal plane, but convex in a transverse plane.

A further embodiment of fixation system is described with reference to Figs. 10 and 11 as regards the system and Fig. 12 as regards its use. Reference numerals correspond generally to those used in Fig. 5(a) for the first embodiment, with 100 added. Again, the system includes epiphysis/metaphysis insert 104 and an implant pin 101. The insert 104 has an angled guide opening 142 through which the implant pin passes.

Additionally, in this embodiment the insert 104 has a corresponding-angled guide opening 146 of small diameter to take a thin plain metal pin e.g. the kind known as a K-wire (Kirschner wire) . K-wires are well known for use in fixing fractures, and are used here in a characteristic way to inhibit rotation of fixated bone. The skilled person will appreciate that small bone screws or pins could be used instead, such as with a tip segment of thread engaging in the femoral head and optionally a head and/or thread at the other end to engage the reaction insert.

The implant pin 101 has an outer element or barrel 102 and an inner element 103 which fits through the barrel 102. The barrel has an enlarged retainer head 121 and a tubular shaft portion 126 extending to an exposed annular cutter edge 129 with forwardly-directed serrations. The shaft region towards the cutting edge at the tip 124 carries several turns of a self-tapping thread 125. In this embodiment the outer element does not comprise any of the expansion or projection structure .

The inner element 103 has an enlarged retainer head 131 with tool engagement formations 139, a smooth cylindrical shaft 136 - which may be tubular for lightness, or may be solid - and a smaller-diameter tip region 134 which carries a thread 135.

A discrete projection structure 127 comprises a pair of channel-form arms 1271 pivoted together at a pivot 1272 with a spring (not shown) urging them apart i.e. to project in opposite radial directions. The pivot includes a small nut (not shown) which threads onto the tip 134 of the inner element 103.

Fig. 11 shows the fully assembled condition including insert and K-wire. It will be seen that, depending on the degree of turning of the inner element 103, the projecting arms may lie in against the small-diameter tip region of the element 103 if held against the spring, whereas if the thread is advanced by tightening the inner element 103 from its head

131, the arms of the structure 127 must then project

outwardly .

For stable abutment against the lateral face of the insert 104, an angled abutment spacer 49 is provided as in the first embodiment. In this embodiment an alternative form of

epiphysis/metaphysis insert 104 is used, which is oblong (in cross-section) in the medial-lateral direction (i.e. generally in the coronal plane of the body, for hip-fracture repair) and the insert 104 is of generally constant thickness top to bottom in the AP direction but downwardly convergent in the medial-lateral direction, so that at the top its cross-section is markedly oblong in the medial direction and its flat side faces 144 at the top are substantially wider than its flat medial and lateral faces 143. Moreover the medial lateral surface 143 is inclined (downwardly convergent) so that in situ (compare Figs. 6 and 11) it will tend to lie on the thicker and stronger bones especially the calcar on the inside of the angle of the proximal femur. Thus, provided that the shape and orientation of the openings formed for the implant pin and insert 104 are precise and accurately aligned, this form of insert although small in size, and not extending down far into the medullary cavity, provides very strong

orientation for the implant pin.

Fig. 12 shows a corresponding procedure. Having examined the fracture - an intra-capsular fracture 912 - by the usual methods, the necessary pin hole 190 and insert opening 195 are formed by the surgeon. They can be formed very accurately with the assistance of good imaging, especially such as described herein. The insert 104 may then be positioned and the K-wire 6 passed through and into the femur head portion for primary holding and prevention of rotation. The outer barrel 102 of the implant pin is then passed through the angular spacer 49 and through the insert 104 (which may have a correspondingly threaded opening 142, or other mechanism to engage the shaft) and turned from its head to engage its outer thread 125 in the bone of the femur neck to the head side of the fracture. An inflatable balloon is then passed through the outer barrel 102 of the pin, either by itself or on the end of a catheter as mentioned above, and inflated with saline to form an enlarged clear void 908 within the femur head in front of the leading edge of the barrel 102. The balloon is then removed, and the inner element 103 passed through the outer barrel. For this, the expandable arms 1271 are held in against their spring to pass through the barrel 102 (Fig.

12 (e) ) but then spring out again when they emerge into the void 908, and turning the inner element 103 to pull the associated nut down the threaded tip 135 consolidates this position. Bone cement is then injected through the inner element 103 (which is hollow, in this variant embodiment) to fill the femur head void 908 and form a solid mass 909 of set bone cement surrounding the expanded arms 1271, enabling the fracture to be impacted as in the first embodiment. It will be noted that the preliminary insertion of the K-wire 6 in this embodiment helps to stabilize the femur head against turning, facilitating safe insertion of the threaded self- tapping barrel 102.

Again, the skilled person will appreciate that this particular mechanism for the expansion of part of the implant pin 101 may be used in other contexts, and with or without a reaction structure insert of the present type. Equally, the reaction structure insert 104 shown therein, which is oblong in the coronal plane, may be used with other forms of implant pin .

The use of a K-wire 6 is just one option. As an

alternative a small bone screw or thin threaded pin may be used, with a thread segment near the tip where it engages in the femur head beyond the fracture, and optionally a formation or mechanism (such as a further thread) for engaging in the insert to apply compressive force.

Fig. 13 shows the use of a third embodiment of implant pin for fixation of a sub-capital fracture. In this

embodiment, no reaction structure (such as an

epiphysis/metaphysis insert, intermedullary device or hip plate) is used, because the bone structure is sufficiently intact to either side of the fracture 912 for the implant pin alone to provide sufficient engagement. The implant pin has an outer barrel 202 rather similar to that of the second embodiment above, having a short external self-tapping thread 225 and a leading cutting edge 229, except that the rear end of the barrel 202 does not have an enlarged retaining edge but only a flat abutment face. The pin design is selected so that the external thread 225 will engage the bone only beyond the fracture line 912. The distinctive feature of this embodiment is the expansion structure of the implant pin, here provided in the form of an inner tubular module or element 203

consisting of a set of metal wires 228 whose forward ends 227 have a rest conformation which is outwardly bent - as shown in Figs. 13(d) and (e) - but which can be held straight against their resilience or shape-memory to fit inside the outer barrel 202 as shown in Figs. 13(b) and (c) .

The inner element comprising the shape-memory wires 228 includes an abutment head 231 which engages against the rear end of the outer tube 202, and an inner stop (not shown) to limit the extent of advance of the wires 228 through the barrel. The wires may be ordinary steel wire or spring leaves held against their natural resilience, or they may be made from a special shape-memory alloy such as nitinol. They can be loaded into an internal tube element relative to which they are slidable, which is then pushed into the outer barrel 202. When ready, the wires are pushed forward through the tubes so that their ends project and extend radially outwards. When embedded in the mass of injected bone cement or other settable compound, they form a strong axial engagement holding the femur head in place. As before, rotational stability may be supplemented by the insertion of one or more additional pins or screws through the fracture. This mode of implant pin may be used with a femoral insert of any kind described herein if wished. No particular mode of injection of the settable compounds such as bone cement is illustrated here. It may be injected either before or after the inner module with the wires is inserted into the outer barrel 202.

Fig. 14 shows stages of fixation of an extracapsular fracture 911, using an implant pin 301 rather similar to that use in Fig. 12, but with a longer segment of external thread 325. The insert 104 is as before. The bone parts are

aligned, the pin hole 190 drilled and insert opening 195 cut as shown in Fig. 14(a) and the insert 104 positioned, Fig. 14 (b) . With this extracapsular fracture 911 the neck and head of the femur are substantially intact, so that the thread 325 of the implant pin 301 can engage securely with the bone beyond the fracture. The engagement is enhanced by expanding the retaining arms 3271 at the tip of the implant pin. In this situation it may not be necessary to inject a mass of set compound (bone cement) in which to embed the pin tip and its arms 3271. If so, an insert pin with an expandable head but without the central channel needed for injecting liquid compound may be adequate. However, in the case of unsound e.g. osteoporotic bone in the femur head, cement may be injected even for this extra-capsular fracture. The injection of smaller quantities of bone cement at the tip and/or at other positions may also be appropriate, although a body or mass is not formed.

Fig. 15 shows a fourth embodiment of implant pin 401. It consists generally of a tubular outer element or barrel 402, a tubular inner element or drive stem 403, and an expansion drive body 405 shown separately in Fig. 15(d) . Unlike the first and second embodiments, this implant pin is for use without a reaction insert extending down from the top of the femur. Instead, it relies on a reaction plate or hip plate 404 also shown in Fig. 15. The reaction plate 404 has an external hip plate portion 441, to lie flat against the lateral outer surface of the femur and having a screw hole 442 for holding it in place, and an angled holding barrel 433 to retain the inner pin element 403. As seen in Figs. 15(b) and (e) the inner element 403 has an enlarged retaining head 431 at its outer end, with a slot for a driving tool, and a screw thread 435 at its tip portion 434. The holding barrel 433 of the reaction plate 404 has an in-turned holding flange 444 defining a hole through which the inner element 403 can pass except for its retaining head 431, so that it can act in conjunction with the reaction plate 404 for tensioning.

The expansion drive body 405 (Fig. 15(d)) has an outward curved cam surface 453 and a threaded bore 452. As in the first embodiment, this threads onto the tip 434 of the drive stem 403 so that when the latter is turned, by a tool acting at its outer end 431, the body 405 is pulled along the stem 403 and spreads the bendable metal limbs 427 which are formed at the end of the outer element 402, separated by slots 423 (Fig. 15(a)). A small holding pin 455 is provided on the outer surface of the expander body 405 to engage between two of the limbs 427 and hold the body 405 against turning with the stem 403.

As before the stem 403 is tubular, and its bore 432 opens at the tip end of the stem as an injection opening 433. The stem 403 fits closely and slidably in the bore 422 of the outer barrel 402.

A further feature enables injection of bone cement at an intermediate region of the implant pin. For this, the stem 403 has a set of outer injection holes 437 towards its outer end. These communicate with corresponding outer injection windows 429 in the outer barrel 402, regardless of the exact rotational orientation (because of peripheral flow grooves seen in Fig. 15(e)) . A further feature in this embodiment - although it may be varied in other embodiments - is that the injection windows 429 are arranged selectively only to one side of the pin 401, so that injected material at this outer region will emerge only on one side of the pin. The reason becomes clear below.

Figs. 16 (a) -(j) show stages of repair of an intra ^¬ capsular fracture using the fourth embodiment of implant pin 401. The procedure uses some variants and additional elements which are not necessarily associated only with the fourth embodiment, but may be used with other types of pin. Equally, the fourth embodiment of pin may be used without all of the present procedural variations.

Fig. 16(a) shows the femur head schematically with an intra-capsular fracture 912 after standard distraction and rotation to restore the correct position. A guide wire 90 is inserted, and in this procedure its entry axis is rather lower than in previous embodiments for reasons explained below. The directing of the guidewire can use the special imaging methods described herein to reduce trial and error so that only a single entry is needed. Using the wire 90 as a guide, a pin hole or bore 190 is drilled using a cannulated drill. Fig. 16(b) shows that the hole lies adjacent the calcar 915. In some variants the drill may remove a small amount from the calcar, shaping it to better complement the implant pin.

A void 498 is then formed in the head of the femur around the pin bore 190. This may be done as in previous

embodiments, using a balloon or the like. However in this embodiment we use a specially-adapted bone crushing device (described later with reference to Figs. 17-19) which is adapted to crush bone only to one side of the axis of

insertion. By this means an eccentric inner void 498 can be formed in the femur head primarily in the sector above, rather than below, the end of the pin bore. The region below the pin bore in the head is not weight-bearing, and the selective formation reduces the amount of setting compound (bone cement) used. Heating on curing is thereby reduced. In Fig. 16(c) the region of the void 498 is shown dotted for illustration.

In the next stage, shown in Figs. 16(d), the bone

crushing tool is used to make a further or intermediate void 499 around the pin bore 190 on the other side of the fracture, above the shaft of the femur. Again, the one-sided action of the bone crushing device enables this void 499 to be formed eccentrically, although all around the pin bore 190, so that its lower part is outward of its upper part and it is formed to a substantial radial extent despite the oblique disposition of the pin bore 190. In Fig. 16(d) the region of the further or intermediate void 499 is shown dotted for illustration.

Fig. 16(e) illustrates a further option, which is also generically applicable in other procedures and with other implant pins contemplated herein. Specifically, an annular occluder 480 is inserted into the bore. This is a tubular component, which fits to extend across the fracture line 912. Its purpose is to prevent bone cement from flowing into the fracture line, where it could hinder healing.

Various options exist for the occluder 480. Generally it will be in the form of a sleeve or tube through which the implant pin can pass or fit, but which will help to block or seal any clearance around the pin. Also, it should be able to hold its position across the fracture. The example shows a helical metal element, in the form of a spring. This is conveniently mountable because it can be tensioned to reduce its diameter for insertion into the bore, and then released to expand at the intended location. However other structures can be contemplated. Indeed, the occluder may be formed as an exterior component of, or formation on, the implant pin itself, and inserted with it. Desirably expansion of the projection elements at the end of the implant pin - here, the limbs 427 - also expands the occluder to enhance its effect.

The implant pin 401, here pre-assembled with the reaction plate 404, is then inserted by the appropriate instrument into the pin bore as shown in Fig. 16(f) . Using the insertion instrument to prevent rotation of the outer element 402 (which has also a tool slot at its outer end 421) the inner element or drive stem 403 is turned by the respective tool formation at its head 431 to draw the driver body 405 back along the thread 435, expanding the limbs 427 into the head void 498 as shown in Fig. 16(g) . This also applies tension through the implant stem 403 between the external reaction plate 404 and the expanded head 427 of the implant outer tube 402, closing the fracture as indicated at 912' . A suitable bone screw 449 is inserted to hold the reaction plate in place - see Fig.

16 (h) .

Low viscosity bone cement is then injected through the central bore 432 of the pin stem from a forward injection position, filling the inner void 498 and curing to form an inner set mass 909' . Further injection of bone cement with a withdrawn position of the injection instrument fills the outer void 499, curing to form an outer set mass 919. The

intervening channels may also fill with the cured compound, as shown in Fig. 16 (j) . The occluder 480 prevents leakage of cement out into the fracture line. By having the pin embedded in an enlarged set mass 909', 919 to either side of the

fracture, an additional reaction effect for tensioning is provided i.e. beyond that provided by the external reaction plate 404. An implant pin without the intermediate injection openings may be used, however. If necessary a larger external reaction plate may be used.

The filling of cement is shown here in two stages, but it could be done in one continuous process. Generally care should be taken to avoid excessive heat emission as the cement cures.

Figs. 17 to 19 show a bone crusher device which can be used for making the eccentric voids mentioned above.

The device has a generally tubular body 601 including a forward extension tube 602 sized to fit into the drilled pin bore. The body has a fixed handle 603 and a pivoted operating lever 604. A shaft 605 is slidable forward inside the body over a predetermined limited stroke against a return spring 646 by actuating the lever 604 around its top pivot 644 above the body; the lever engages the shaft 605 at a link pivot 645.

The tube 602 has a crushing tip 655 with a moving plate

(crusher shoe) 607 connected to a static plate 606 by a linkage 608 whereby the crusher shoe 607 moves in and out radially and with its outer face remaining substantially parallel to the tube 602. The static plate is provided by an extension of one side of the tube 602. The linkage has a forwardly-inclined rear link 688 connecting at a rear pivot pin 678 of the crusher shoe and a pair of parallel rearwardly- inclined front links 689 connecting respectively to the rear and front pivot pins 678,679 of the crusher shoe to assure a stable action parallel to the tube 602 and substantially radial, as shown by arrow B in Fig. 19(a) . The user can shape an eccentric void by adjusting the depth of insertion - by known means - and the orientation of the handle 603, knowing that crushing is selectively in the direction opposite to the handle. The static plate 606 is aligned with the main wall of tube 602 which is supported all along the other side of the pin bore 190, assuring selective crushing at the side of the crusher shoe 607 only.

Finally, Figs. 20 and 21 illustrate an improved way of implementing radiography during surgery so that the surgery can be more quickly and easily carried out and in particular accurately because of better visualization of the surgical site without obstruction of the surgical activity.

Figs. 20 and 21 illustrate imaging axes for a novel imaging method proposed herein as used in hip surgery.

Letters A, B and C indicate individual imaging axes for radiography, in particular fluoroscopy during surgery. A is an anterior/posterior axis. B indicates a second imaging axis which is perpendicular to the femur neck but not orthogonal to the AP axis A. Rather, it is angled obliquely to the first axis A at an angle "x" which may be e.g. between 60° and 70°.

Desirably a third imaging axis C is also used, also angled obliquely relative to the first axis at an angle "x" (which may be the same as or different from that for the second axis B, but desirably in the same range) , and viewing perpendicular to the shaft of the femur. By imaging along the oblique axes

B and C the detectors of the corresponding imaging apparatus can be positioned down below the level of the surgical table 150 to reduce obstruction of the surgical site. Although axes C and B are not even close to lateral, their (second and third) image data can be taken in combination with the image data from the first imaging axis A and, by means of image analysis processing, converted to one or more reconstructed image data sets which do correspond to a direct lateral view (or indeed, any other direction of view that the surgeon may wish to use) or data sets taken from a library of images or reconstructions. These images may then be displayed e.g. in the conventional manner, on a pair of screens, one showing the AP view and the other the lateral view.

It will be noted that, while the present descriptions sometimes refer to the personal activity of a surgeon in manipulating implant pins, drilling openings and so forth, this personal activity is not necessarily intrinsic to the methods and apparatus described herein which can also be useful with automated, partly-automated or remotely-controlled surgical methods.