Radial Head Fracture- Open Reduction and Internal Fixation with Medartis 3.0mm CCS Screws and LUCL repair using Arthrex Biocorkscrew anchor

AP radiograph of left elbow
36 year old right handed manual worker sustained this closed injury after a fall.
He presented with a swollen, painful elbow with limited range of movement.
This image shows the displaced radial head fracture (A). It appear to still be articulating with the capitellum but there is clear displacement of the radial head from the neck.
The ulnohumeral articulation (B) appears to be congruent.

Lateral radiograph of left elbow
This radiograph shows the ulnohumeral articulation is in joint.
The radial neck (B) is aligned with the capitellum (C) but the head fragment (A) appears to lie posterior (with the eye of faith), but is not clearly seen.

CT Imaging – Coronal view
The head fragment (A) has subluxed posteriorly and laterally
It is still partially articulating with the capitellum (C).
B – olecranon

CT imaging – sagittal view
This view shows the radial head fragment (A) has displaced posteriorly in relation to the radial neck (B)
It is still partially articulating with the capitellum (C).

CT imaging – 3D reconstruction
This image clarifies the pattern of the fracture and the relation with the anatomy of the elbow as described in the previous images.

The patient is set up supine on the lateral edge of the table, with the arm on an arm table, prepped and draped.It is important to ensure that the body is positioned close to the edge of the table. This allows adequate intra-operative fluoroscopy images.
A high arm tourniquet is applied (A) before prepping and draping.
A gown pack can also be placed under the elbow to aid intra-operative visualisation.

With the elbow flexed, the incision is sited 2-3cm proximal to the lateral epicondyle and extends distally to the posterior border of the ulna at a point 10cm distal to the tip of the olecranon (D).The landmarks include:
lateral epicondyle (A)
olecranon (B)
and radial head (localised on pronation/supination of forearm) (C)
The distal extent of the incision is important to help define the anatomy and particularly the intermuscular plane.

Once an incision through the skin and fat is made, a West retractor is used to maintain tension when dissecting through the superficial soft tissues.

In acute fractures, the soft tissues can be very contused as shown in this image.

The deeper soft tissues are dissected down to the fascial layer using a 15 surgical blade or McIndoe scissors and the fat is swept off the muscle fascia to define anatomy.The subcutaneous fat is swept off the fascia to help define the muscle planes for the approach.
If greater exposure is required, the incision can be extended proximally along the lateral intermuscular septum.

Once the fascia has been defined, the bony landmarks can be palpated to confirm the anatomy.A – anconeus fascia
B – ECU fascia
C – location of displaced radial head fracture
Kocher’s approach is in the muscle plane between ECU and anconeus and is clearly defined by the dip between the 2 muscle bellies (D).
It is not always so clear as the plane between the 2 muscle bellies can be ill defined.
It is best identified by looking at the orientation of the muscle fibres and also by a dip in the convexity of the muscle fascia due the tethering of the deep fascia between the two compartments (D). The fibres themselves run longitudinally in both muscles but diverge very subtly. This is usually better seen when the fascia is incised.

The fascia between anconeus and ECU is incised and is extended towards the distal extent of the incision. Proximally, it is extended up to the lateral epicondyle and along the lateral intermuscular septum for 2-3cm

The muscle bellies of anconeus and ECU are retracted using the West retractor.A Kocher’s approach utilises the intermuscular plane between anconeus (radial nerve) and extensor carpi ulnaris (posterior interosseous nerve, PIN).
At the level of the lateral epicondyle (A), the muscles of the extensor origin are sequentially released proximally to aid exposure. This is done by extending the incision proximally from between anconeus and ECU up to the lateral epicondyle and along the lateral ridge. The The extensors are dissected off the bone and retracted anteriorly.
The attachments of the extensor origin from proximal to distal are:
Extensor carpi radialis longus (ECRL) (B)
Extensor carpi radialis brevis (ECRB) (C)
Extensor digitorum communis (EDC) (D)
Extensor carpi ulnaris (ECU) (E)
The muscles of the extensor origin attach onto the lateral epicondyle and the lateral supracondylar ridge. Note, it is not easy to isolate each muscle attachment as they tend to blend together.

The humeral attachment and competency of the lateral collateral ligament (LCL) is assessed.The LCL complex comprises of the radial collateral ligament, annular ligament, lateral ulnar collateral ligament (LUCL) and accessory lateral collateral ligament.
The humeral attachment of the LUCL is marked by the forceps in this image. It has been avulsed off the lateral epicondyle (A).
Fractured radial head as shown (B).
The LUCL originates from the lateral epicondyle of the humerus (A) and attaches to the supinator crest of the ulna (C). Its function is resist varus and external rotation forces. Incompetence of this can lead to posterolateral rotational instability.

The joint capsule and annular ligament is incised distally to expose the radial neck.Fractured radial head – B
The incision into the capsule and annular ligament is made in the same orientation as the incision for Kocher’s approach. It runs from the lateral epicondyle towards the radial neck.
At the level just distal to the lateral epicondyle, the capsule is incised. As the incision is extended distally, the capsule condenses to form the annular ligament (A) at the proximal edge of the radial head. The annular ligament is a strong ligament that winds round the radial head and neck and attaches at the anterior and posterior margins of the radial notch of the ulna. It stabilises the radial head within the elbow joint and in its relation to the proximal ulna, whilst allowing pronosupination.
A radial incision is made into the annular ligament as shown to expose the radial neck.
There are pitfalls when incising at this layer. If a line is drawn from C – D:
Care is taken not to stray posteriorly as this can cause an iatrogenic injury to an intact LUCL.
Care is also taken not to stray too anteriorly as the posterior interosseous nerve (PIN) can be injured.
Forearm pronation minimises the risk of injury to the PIN but moving the course of the nerve away from the surgical approach.
The PIN is a continuation of the radial nerve after it divides from the superficial sensory branch in the arm and is also known as the deep branch of the radial nerve. Proximally, the PIN is separated from the joint by the brachialis muscle. As it courses distally, it passes under the proximal edge of supinator muscle (E) under the Arcade of Frohse before continuing on the dorsal surface of the interosseous membrane.
In this image, the transverse fibres of supinator are seen just distal to the annular ligament.
The Arcade of Frohse is the most proximal part of the superficial layer of the supinator muscle, and is a fibrous arch over which, the posterior interosseous nerve passes.

Distally, the proximal fibres of supinator may be split to aid exposure of the neck. This is done by extending the incision distal to the annular ligament into the proximal transverse fibres of supinator. If this step is required, it is common to head more posteriorly to avaoid injuring the posterior interosseous nerve.
It is of particular importance in this step that the forearm remains in pronation and approach the lateral aspect of the radial neck to minimise risk of injury to the PIN.
A – proximal edge of supinator muscle

The elbow joint is washed out to clear any haematoma/haemarthrosis and to remove any loose fracture fragments using normal saline in a 50ml bladder syringe.

The radial head is delivered by rotating it into view and is inspected for any residual soft tissue attachment. This is best done by manually manipulating the radial head using forceps or fingers. By using (very) gentle traction, it is possible to assess for any remaining soft tissue attachment to the fractured radial head/neck.
Any damage to the chondral cartilage can also be documented.

By manually manipulating the radial head fracture out (and respecting any soft tissue attachments), the elbow joint can be exposed and any loose bodies can removed from the elbow joint using forceps and further washout..In this image, the head been gently displaced distally to allow joint visualisation. Care is taken to respect any remaining soft tissue attachment in this manoeuvre.

The joint is inspected to assess for chondral damage to the coronoid and capitellum.A – radial head
B – capitellum
This image also shows that there is still some (or very little) capsular attachment to the radial head (C).

In this image, the anterolateral facet of the coronoid is visualised (C). It is important to note any associated fractures of the anterolateral facet of the coronoid, although the lead author is inclined to manage these fractures conservatively if they involve less than 50% of the coronoid height and the elbow is stable once other pathology has been adressed.
Note that due to the radial head fracture and LCL injury, the elbow is unstable and has subsequently dislocated posteriorly at its ulno-humeral articulation.
This is shown by the high position of the capitellum (B) in relation to the radial head (A).

If required, the elbow joint is reduced by traction of the forearm and flexion of the elbow.
In simple terms, this can be done by elevating the patient’s hand and flexing the elbow. The weight of the arm usually provides enough countertraction to reduce the elbow’s ulnohumeral articulation.
This is usually confirmed by a clunk back into position and the image shows that the capitellum (A) is aligned with the radial head (B).

The radial head is held by the finger and thumb and reduced back into position by pushing it back into joint on the radial neck. If there is resistance, this can be aided by a varus stress as well as pronosupination to optimise the space for reduction.In this image, the radial head (A) is aligned with the neck (B) and the capitellum (C).
Note the slight rotational malalignment of the radial head fracture with the radial neck fracture pattern.

The rotational alignment can be fine-tuned using a Hawarth periosteal elevator (A) and/or a Macdonald retractor (B). The instruments are inserted into the fracture site and rotated to distract fracture. The forearm is pronosupinated until a satisfactory position is achieved. The instruments are then removed to allow contact at the fracture site.

The reduction is usually accepted when the fracture edges are lined up as shown in this image.
In a young patient where the bone quality is good and little comminution, you can get a good cortical read and get an anatomical reduction.
Where there is comminution, there has to be some extrapolation of where those fragments sit within the fracture pattern. There is usually enough information to get a near anatomical reduction. If this is not possible due to significant comminution, it is fair to question whether fixation is appropriate as there may not be enough inherent stability once reduced. Also, any malreduction can affect the pronosupination arc and overall range of movement. If there is doubt, a radial head replacement should be considered.

A 1.1mm Kirschner wire (K-wire) is sited using a wire driver. the wire runs from the margin of the radial head, and away from the articular surface (as shown). It is driven in at an angle of ~45 degrees until it passess through the cortex of the radial neck with the forearm in neutral rotation.There should be feedback as the wire is passed through the subchondral bone, cancellous bone and the cortical one of the radial neck.
A Howarth or Macdonald placed around the radial neck to define it’s dimensions can aid spatial awareness when siting the K-wire.
Care is taken to stop advancing once the wire is through the far cortex of the radial neck.
This will allow accurate measurement of the screw length.
If there is any doubt, this can be checked under image intensifier, usually after a second wire is passed to ensure the reduction is stable.
The lead surgeon does not routinely image if satisfied that there is good feedback when inserting the K-wire. This is to minimise the risk of displacement when manipulating the elbow into a suitable position for fluoroscopic imaging.
For radial neck fractures, it is not possible to maintain reduction using pointed reduction clamps. For displaced partial radial head fractures, it is possible to reduced the fracture and place a pointed reduction clamp to maintain reduction. However, this can be technically difficult due to the limited space. The lead author tends to prefer using K-wires to maintain reduction.

The forearm is then rotated into maximal supination as far as the original K-wire position allows.In this image, the K-wire has now moved to a more posterior position with supination of the forearm.

A second 1.1mm K-wire is inserted from the radial head into the radial neck form an anterior position using the same principles as for the first K-wire.The aim is to allow the placement of 3 screws running from the radial head and and engaging into the cortex of the radial neck.
As the screws all converge into a limited space on the radial neck, some forward planning is required to ensure 3 screws can be sited from the radial head into the radial neck, i.e. the screws should be aimed slightly eccentrically to allow passage of all screws.

The size of both screws is referenced off the length of the K-wires using the depth gauge.For the measurement to be accurate, it is important that the K-wire is only just advanced past the far cortex.
It is common for the lead surgeon to take 2-4mm off the screw length to allow adequate compression of the fracture without the screw becoming prominent either proximally or distally.

A 2.1mm cannulated drill is used to drill over the K-wire from the cortex of the radial head, through the cancellous bone and through the cortex of the radial neck.Note that this step is optional. It is possible to skip this step in osteopaenic bone or if there is concern regarding maintaining reduction.
The K-wire is completely overdrilled.
When drilling, check that the drill is running in the same alignment as the K-wire. A drill sleeve can be used to maintain the direction of the K-wire and protect the soft tissues (not used in this image).
If the alignment is incorrect, the K-wire may be notched and can break in situ if drilling is forced.

When the K-wire is overdrilled, the K-wire may back out when removing the drill, as is shown in this image.

If the K-wire has backed out when drilling, the K-wire is replaced and can be done so by hand. If there is concern that it has not followed the same track, it should be checked under image intensifier.

Note all the instruments are colour coded. For 3.0mm headless screws, the drills should have a yellow mark as shown.
In this image, the yellow marked countersink is loaded on the driver and is used for overdrilling near cortex.

The near cortex only is overdrilled using the countersink.Optional step. This is used if the bone quality is good.
The near cortex is overdrilled to allow engagement of the threads in the proximal part of the headless screw.
During the steps to drill and countersink, it is important to retract the soft tissue out of the way so that it does not get caught up in the drill.

An image of the 3.0mm CCS screw loaded on the appropriate star drive screwdriver.
Note the variable pitch thread on the shaft (A) and head (B) that allow compression of the fracture once the head engages in the near cortex.

The 3.0mm CCS screw is loaded onto the screwdriver and passed over the K-wire and fixed across the fracture.

The screw is advanced into position so that there should be feedback that the distal screw threads have engaged in the cortex of the radial neck and the proximal screw thread in the radial head. The head of the screw should be buried deeper than the height of the cartilage to ensure that it doesn’t irritate the chondral cartilage in the joint or cause crepitus.

The K-wire is be removed by hand.

The second K-wire is then drilled and prepared as as per the same sequence for the first screw so that the second headless screw can be inserted.This includes the steps described from slide 30-36.

The second screw is inserted and K-wire removed by hand. This frees up the elbow of K-wires to allow maximal pronation to expose the entry point for the third K-wire. The third wire is inserted using the same principles as the first 2 K-wires, running from the radial head into the cortex of the radial neck at an angle of ~45 degrees.
Note in this image, the visible headless screw (1) on view is the first screw that we inserted.
As the forearm is now positioned in maximal pronation, screw 1 has now rotated on the radial head into an anterior position and screw 2 (not visible on the image) has rotated medially towards the proximal radial notch of the ulna.
This exposes the posterior aspect of the radial for siting of the third screw (3).
This step can only be performed once 2 screws have been applied across the fracture site to confer some stability. The K-wire have to be removed to allow the radial head to rotate to bring the access point into view for the third screw.
Despite having 2 screws in situ, this image also shows that there has been some loss of reduction. This as accepted by the lead author as the arc of movement appeared smooth and free of crepitus. The fracture also appeared to be stable in this position.

The third K-wire is measured, (drilled) and the third headless screw is inserted over the K-wire and advanced from the radial head into the radial neck.See slides 28-38 for tips on this aspect of the technique.

Once the screws have been inserted, the elbow is taken through a full range of movement in flexion and extension to ensure there is no crepitus, as well as being screenedCrepitus can occur either from:
Prominent screw heads
Prominent screw tips
Inadequate reduction of the fracture
These tend to catch as it articulates with the ulna at the proximal radial notch of the ulna.

The fixation is screened under II in two planes to check adequacy of fixation and stability. The elbow is typically put through a full range of movement in flexion and extension and pronosupination to ensure a congruent arc of movement.
This fluoroscopic image shows an AP view of the elbow. It shows that the screws are engaged in the cortex of the radial neck and not too prominent or in a position to irritate the elbow joint surfaces.

This lateral view of the elbow shows the screws in an appropriate position in 2 planes.
There is some residual subluxation of the ulnohumeral articulation as the LUCL has not yet been repaired.
This can be seen as the increased space between the distal humerus and the articulation of the olecranon (A).

The incised annular ligament is identified.
A – anterior leaflet of annular ligament
B – posterior leaflet of annular ligament (held in forceps)

The annular ligament is repaired using simple interrupted 1/0 vicryl sutures starting from distal and working proximally. The knots are laid on the superficial aspect of the annular ligament to minimise irritation on the radial neck.

The annular ligament is usually repaired using 2 interrupted sutures.

If the LUCL is avulsed from its humeral attachment, attention is turned to its repair at this stage.
In this injury, the LUCL origin (A) is noted to have avulsed off the lateral epicondyle (B).

When the humeral attachment the LUCL is restored, the sling effect of the LUCL on the radial head is recreated.
By recreating this, the primary stabiliser to varus stress is restored. It also stabilises the radial head and provides a restraint to stop the radial head subluxing posteriorly and laterally resulting in posterolateral rotatory instability (PLRI).
The primary stabilisers of the elbow are:
Ulnohumeral articulation
Medial collateral ligament
Lateral collateral ligament complex

The humeral attachment of the LUCL is identified on the lateral epicondyle at the centre of the radius of the capitellum.A – lateral epicondyle.
The centre of rotation is defined by the centre of the radius of the capitellum.
In this image, the forceps are used to demarcate the posterior aspect of the capitellum to help accurately locate this position (B).

This image show the forceps being used to accurately note the anterior aspect of the capitellum (A).

The centre of rotation on the lateral epicondyle is marked using the diathermy.A – centre of rotation

The tap for the Arthrex 3.7mm Biocorkscrew anchors is used to prepare the bone tunnel.
Note the laser line on the tape (A).

The Biocorkscrew tap is placed on the lateral epicondyle as previously denoted by the diathermy mark. The mallet is used to engage the tip of the tap into the bone.

The tap is advanced by turning in a clockwise manner down to the laser line on the tap. The tap is aimed towards the distal aspect of the medial epicondyle, in line wit the axis of rotation for the elbow.This is done under direct vision and is to ensure that the tap does not exit into the joint and thereby damaging the chondral cartilage.
There is a margin of error of around 1.5cm in each direction anteriorly, posteriorly and distally.
The tap has to be advanced 2-3cm down to the laser line.
If in doubt, a finger can be placed anteriorly on the joint surface over the capitellum and trochlear to palpate the margins and provide a guide for tap direction.

An Arthrex 3.7mm Biocorkscrew anchor double loaded with Fibrewire sutures with curved needles

The Arthrex 3.7mm Biocorkscrew anchor is advanced into the bone tunnel by turning clockwise down to the laser marking on the introducer.

The lid is slid off the handle to expose the Fibrewire sutures.

The Fibrewire sutures with attached needles are freed from the handle.

The handle is pulled away to undock the Biocorkscrew anchor, leaving it deployed within the bone.

The needles of the Fibrewire sutures are unloaded from the handle by passing them through the fenestration at the distal aspect of the handle (A).

The 4 limbs of Fibrewire suture are passed through the avulsed humeral attachment of the LUCL (A).

Each pair of sutures is identified and isolated. The first pair of sutures are tied using a mattress suture is tied to secure the LUCL onto the lateral epicondyle.

One limb of the second Fibrewire suture is used to Whipp stitch the LUCL to provide extra security.The Whipp stitch only requires 3 throws due to limited available tissue.
The Whipp stitch is run for 3 throws along the anterior part of the avulsed LUCL, running from proximal to distal. It is then run up the posterior aspect of the avulsed LUCL from distal to proximal.

The suture is slid down bu pulling on the other limb of the second suture to reduce the avulsed LUCL so that there is bony contact with the LUCL and lateral epicondyle. The suture is tied and secured with 3-4 locking throws and cut.

Image of repaired LUCL ligament (C).
A – anconeus fascia
B – ECU muscle
D – Extensor muscles
E – repaired annular ligament

The muscle fascia layer between anconeus and ECU distally and muscles of the extensor origin proximally are repaired using 1-0 vicryl.A continuous running suture is used.

The fat layer is closed using 2-0 vicryl and 3-0 monocryl for the skin.

1/2 inch are steristrips applied.
A waterproof dressing and wool and crepe bandages are applied.

This is an early AP/oblique radiograph of the elbow post fixation. This fixation is holding well.
Although the most superior screw (A) looks prominent, this is likely to be projectional and not likely to be irritating the joint either way.
As noted from the intra-operative reduction, the fracture is not quite anatomically reduced at the radial neck (B), but likely due to some comminution and accpetance of the final position intra-operatively. It is not likely to be causing a clinical problem.
Unfortunately, the patient did not attend subsequent appointments.

A lateral radiograph showing the position is maintained.
Although the anterior screw looks prominent, this is again likely to be projectional as the x-ray beam would need to be orthogonal to the screw to get a true appreciation as to the screws precise location.
As it sits at the margin of the radial head, it is not likely to cause irritation at the radio-capitellar articulation even if fractionally prominent.
If the screw does irritate the proximal radial notch of the ulna, there is likely associated crepitus and pain.
Note that over time, screws can occasionally back out 1-2mm and irritate the joint surface. It is important to be open to this problem as the metalwork may need to be removed once the bone has healed.