Setup
Under general anesthesia, the patient is placed in a lateral position with
the arm on a bolster. A sterile tourniquet is applied to the upper arm and is
inflated to 250 mm Hg.
Exposure
A global approach, which is a 20 to 25-cm posterior midline incision
centered on the olecranon, is used. Full-thickness fasciocutaneous flaps are
elevated12. The
Kocher interval between the anconeus and the extensor carpi ulnaris is
identified by a thin strip of fat, which can be seen deep to the deep
fascia13 (Figs.
3-A and
3-B). The deep fascia at this
interval is incised, and the muscles are retracted to expose the lateral
ligament complex. Care is required to dissect out the capsule from the
muscular layer. This is an important step as the ligamentous-capsular
structures are the primary stabilizers of the joint and, if torn, need to be
properly repaired at the completion of the procedure to maintain
stability.
"Z" Lateral Capsulotomy
The use of a z-shaped capsulotomy is designed so that stability is
compromised only in those patients in whom the extra exposure is required. It
ensures that the surgeon can obtain a stable repair of the annular ligament
following insertion of a metallic prosthesis. An incision is made in the
capsule anterior to the lateral ulnar collateral ligament. We recommend a
z-shaped incision through the annular ligament as this ensures that a strong
repair can be
achieved11,14
(Figs. 4-A and
4-B). Anterior and posterior
capsular flaps are created. The anterior capsular flap can be
released from the epicondyle, exposing the anterolateral aspect of the joint
without violating the lateral ulnar collateral ligament
(Fig. 4-C). At this stage, the
posterior capsular flap remains intact to maintain joint stability. A radial
head fracture can be visualized well with this limited approach, and open
reduction and internal fixation can be performed. With a delayed procedure,
the patient may have a fixed flexion deformity, which can be corrected by
releasing the capsule from the anterior aspect of the humerus. Simple
soft-tissue closure is all that is required if only this limited anterior
capsular release has been performed.
Releasing the posterior capsular flap disables the lateral ulnar
collateral ligament and therefore makes the elbow unstable
(Fig. 4-D). It is indicated if
extra exposure is required to achieve an open reduction and internal fixation
of a radial head fracture or for the insertion of a metallic prosthesis. If
the posterior capsular flap is released, then a strong repair is required,
usually with use of transosseous number-1 or 0 Ethibond sutures (Ethicon,
Johnson and Johnson, Somerville, New Jersey).
Assessment of Elbow Stability
If stable fixation of the radial head fracture cannot be obtained, then the
radial head is excised. The medial collateral ligament and interosseous
radioulnar membrane are then assessed for stability. Medial collateral
ligament stability is assessed by placing a valgus force on the elbow and
checking the distance between the radial neck and the capitellum
(Fig. 5-A). The forearm is
placed in pronation with the elbow at 30° of flexion. Narrowing of >2
mm in the distance between the radial neck and the capitellum when a valgus
stress is applied is indicative of a disruption of the anterior band of the
medial collateral ligament. If the elbow is also unstable in the extended
position, it indicates that the anterior and posterior capsule is also
torn11.
A Kocher clamp is then placed on the radial neck, and the radius is axially
loaded to assess the interosseous membrane competence. A change of >2 mm in
the distance between the radial neck and the capitellum is abnormal
(Fig. 5-B). Both of these
stress tests can be performed under fluoroscopic control.
It has been our experience that the lateral ligament complex is avulsed
from the humerus as a complete ligamentous sheet. In patients who had a
delayed procedure, we noted that the lateral capsular complex had often not
healed and had been displaced distally, overlying the capitellum, to which it
cannot heal. We believe that this is likely to be the cause of posterolateral
rotatory instability.
CRITICAL CONCEPTSINDICATIONS:The indications for use of the metallic radial head prosthesis include: (1)
an acute comminuted fracture of the radial head in which satisfactory
reduction and/or fixation cannot be obtained, such that early motion could be
instituted; (2) delayed presentation of radial head fractures with persistent
pain or instability; and (3) patients with complex elbow injuries that include
a rim fracture representing >30% of the radial head, which cannot be
reconstructed. Inserting a radial head prosthesis in these patients is much
better than leaving a sizeable rim fragment, which is likely to lead to
recurrent instability.CONTRAINDICATIONS:Active infectionPITFALLS:The main pitfall of the procedure is to underestimate the importance of the
stability of the elbow. A comminuted fracture of the radial head is part of
the spectrum of elbow instability, and there is a high likelihood of
associated ligamentous injuries. It is imperative that the surgeon respect
these associated injuries to prevent persistent instability. It is our opinion
that fixation of the coronoid process provides the greatest improvement in
stability of the elbow. Repair of the disrupted medial and lateral ligament
complexes is also important.The metallic prosthesis that is inserted should replicate, as closely as
possible, and correspond in size to the native radial head. The new modular
prostheses, EVOLVE radial head (Wright Medical Technology, Arlington,
Tennessee) and rHead radial implant system (Avanta Orthopaedics, San Diego,
California), provide greater flexibility, enabling the surgeon to more closely
reproduce the normal anatomy (Fig.
10). A prosthesis with a diameter that is too large will point
load on the margins of the sigmoid notch, whereas a prosthesis that is too
small will point load on the sigmoid notch. A radial head with an incorrect
diameter also has a cam effect, which produces abnormal loading on the
capitellum.The height of the prosthesis is also important. Fortunately, most fractures
occur at the head-neck junction, and, once the head is excised, insertion of a
prosthesis that corresponds in size to a native head will be adequate. It is
important to ensure that the radial head is loading on the capitellum and is
not "overstuffed," which may lead to pain on the lateral side of
the elbow. Insertion of a radial head that is too short, such that it does not
load on the capitellum, increases the instability of the joint.A patient who presents with persistent problems after having an acute
radial head excision can be a major surgical challenge. These patients often
present with pain, a sensation of instability, and radiographs that
demonstrate valgus deformity of the elbow with opening of the medial joint
space. These are particularly difficult cases, as the soft tissues on the
lateral side of the elbow have shortened following the radial head excision,
the radial nerve is not in its normal anatomical position, and the medial
collateral ligament is usually incompetent. Finally, at surgery, there is no
simple reference as to the appropriate diameter or height of the metallic
prosthesis. The height of the prosthesis should be to the level of the
trochlear groove, and the diameter should correspond to the curvature of the
sigmoid notch. Tertiary referral of these cases would seem to be
appropriate.Fractures of the radial head that extend into the neck can be difficult to
manage. It is important that the surgeon carefully inspect the neck of the
radius to identify any fractures that may be undisplaced. We recommend that
undisplaced fractures be prophylactically wired prior to insertion of the
metallic prosthesis to prevent displacement when the prosthesis is
inserted.While there may be a rare indication to cement the prosthesis in the
management of tumors, it does not need to be cemented in a posttraumatic
situation. The stem of the EVOLVE prosthesis is polished and circular in cross
section. This allows the stem to rotate within the medullary canal of the
proximal part of the radius, effectively creating a bipolar prosthesis. It can
rotate at the radiocapitellar articulation and between the stem and the neck
of the proximal part of the radius.We have seen two patients who had "proximal radial escape."
This occurs with excessive excision of the radial head and proximal aspect of
the neck, such that the proximal part of the radius is no longer contained by
the annular ligament. The proximal part of the radius is then pulled in an
anterior direction by the intact biceps tendon. When the patient attempts to
supinate against resistance, the proximal part of the radius is pulled forward
into the cubital fossa, producing discomfort with minimal improvement in
supination strength. Radial head escape can be avoided, ensuring the integrity
of the annular ligament, by avoiding excessive excision of the radial head and
neck.AUTHOR UPDATE:The major advance in the management of comminuted fractures of the radial
head in the last few years has been the development of modular prostheses.
They provide many permutations in the size of the head and shaft components,
allowing the surgeon to more closely individualize the prosthesis to the size
and shape of the native radial head. The EVOLVE radial head (Wright Medical
Technology) and rHead radial head (Avanta Orthopaedics) are examples of this
new generation of modular prostheses. Plates with fixed-angle screws are also
an important advance and allowsuperior fixation of less severe fractures,
which previously were managed with replacement. We believe that the radial
head prosthesis should be limited to the head and neck region only. Long
prostheses, which extend distal to the 15° angulation of the proximal part
of the radius, are likely to create an angular offset from the native radial
head with forearm rotation. These long prostheses should be avoided when
possible. However, there are cases in which a longer prosthesis is required
because of proximal fracture extension or a previous extensive excision. We
believe that this is the ideal situation for an articulated prosthesis, as the
articulation minimizes the offset problems of a prosthesis that extends past
the proximal radial angulation.We recently had one patient in whom a number-5 Ethibond suture that was
used to fix the coronoid process fracture frayed and ruptured, resulting in a
recurrent dislocation. Thus, we now use number-5 FiberWire as it is less
likely to fray.For most patients with an unstable elbow and for those with a delayed
presentation, dynamic external fixators and ligament grafts have a place in
providing additional joint stability.
CRITICAL CONCEPTS
INDICATIONS:
The indications for use of the metallic radial head prosthesis include: (1)
an acute comminuted fracture of the radial head in which satisfactory
reduction and/or fixation cannot be obtained, such that early motion could be
instituted; (2) delayed presentation of radial head fractures with persistent
pain or instability; and (3) patients with complex elbow injuries that include
a rim fracture representing >30% of the radial head, which cannot be
reconstructed. Inserting a radial head prosthesis in these patients is much
better than leaving a sizeable rim fragment, which is likely to lead to
recurrent instability.
CONTRAINDICATIONS:
Active infection
PITFALLS:
The main pitfall of the procedure is to underestimate the importance of the
stability of the elbow. A comminuted fracture of the radial head is part of
the spectrum of elbow instability, and there is a high likelihood of
associated ligamentous injuries. It is imperative that the surgeon respect
these associated injuries to prevent persistent instability. It is our opinion
that fixation of the coronoid process provides the greatest improvement in
stability of the elbow. Repair of the disrupted medial and lateral ligament
complexes is also important.
The metallic prosthesis that is inserted should replicate, as closely as
possible, and correspond in size to the native radial head. The new modular
prostheses, EVOLVE radial head (Wright Medical Technology, Arlington,
Tennessee) and rHead radial implant system (Avanta Orthopaedics, San Diego,
California), provide greater flexibility, enabling the surgeon to more closely
reproduce the normal anatomy (Fig.
10). A prosthesis with a diameter that is too large will point
load on the margins of the sigmoid notch, whereas a prosthesis that is too
small will point load on the sigmoid notch. A radial head with an incorrect
diameter also has a cam effect, which produces abnormal loading on the
capitellum.
The height of the prosthesis is also important. Fortunately, most fractures
occur at the head-neck junction, and, once the head is excised, insertion of a
prosthesis that corresponds in size to a native head will be adequate. It is
important to ensure that the radial head is loading on the capitellum and is
not "overstuffed," which may lead to pain on the lateral side of
the elbow. Insertion of a radial head that is too short, such that it does not
load on the capitellum, increases the instability of the joint.
A patient who presents with persistent problems after having an acute
radial head excision can be a major surgical challenge. These patients often
present with pain, a sensation of instability, and radiographs that
demonstrate valgus deformity of the elbow with opening of the medial joint
space. These are particularly difficult cases, as the soft tissues on the
lateral side of the elbow have shortened following the radial head excision,
the radial nerve is not in its normal anatomical position, and the medial
collateral ligament is usually incompetent. Finally, at surgery, there is no
simple reference as to the appropriate diameter or height of the metallic
prosthesis. The height of the prosthesis should be to the level of the
trochlear groove, and the diameter should correspond to the curvature of the
sigmoid notch. Tertiary referral of these cases would seem to be
appropriate.
Fractures of the radial head that extend into the neck can be difficult to
manage. It is important that the surgeon carefully inspect the neck of the
radius to identify any fractures that may be undisplaced. We recommend that
undisplaced fractures be prophylactically wired prior to insertion of the
metallic prosthesis to prevent displacement when the prosthesis is
inserted.
While there may be a rare indication to cement the prosthesis in the
management of tumors, it does not need to be cemented in a posttraumatic
situation. The stem of the EVOLVE prosthesis is polished and circular in cross
section. This allows the stem to rotate within the medullary canal of the
proximal part of the radius, effectively creating a bipolar prosthesis. It can
rotate at the radiocapitellar articulation and between the stem and the neck
of the proximal part of the radius.
We have seen two patients who had "proximal radial escape."
This occurs with excessive excision of the radial head and proximal aspect of
the neck, such that the proximal part of the radius is no longer contained by
the annular ligament. The proximal part of the radius is then pulled in an
anterior direction by the intact biceps tendon. When the patient attempts to
supinate against resistance, the proximal part of the radius is pulled forward
into the cubital fossa, producing discomfort with minimal improvement in
supination strength. Radial head escape can be avoided, ensuring the integrity
of the annular ligament, by avoiding excessive excision of the radial head and
neck.
AUTHOR UPDATE:
The major advance in the management of comminuted fractures of the radial
head in the last few years has been the development of modular prostheses.
They provide many permutations in the size of the head and shaft components,
allowing the surgeon to more closely individualize the prosthesis to the size
and shape of the native radial head. The EVOLVE radial head (Wright Medical
Technology) and rHead radial head (Avanta Orthopaedics) are examples of this
new generation of modular prostheses. Plates with fixed-angle screws are also
an important advance and allowsuperior fixation of less severe fractures,
which previously were managed with replacement. We believe that the radial
head prosthesis should be limited to the head and neck region only. Long
prostheses, which extend distal to the 15° angulation of the proximal part
of the radius, are likely to create an angular offset from the native radial
head with forearm rotation. These long prostheses should be avoided when
possible. However, there are cases in which a longer prosthesis is required
because of proximal fracture extension or a previous extensive excision. We
believe that this is the ideal situation for an articulated prosthesis, as the
articulation minimizes the offset problems of a prosthesis that extends past
the proximal radial angulation.
We recently had one patient in whom a number-5 Ethibond suture that was
used to fix the coronoid process fracture frayed and ruptured, resulting in a
recurrent dislocation. Thus, we now use number-5 FiberWire as it is less
likely to fray.
For most patients with an unstable elbow and for those with a delayed
presentation, dynamic external fixators and ligament grafts have a place in
providing additional joint stability.
Medial Exposure
The medial side of the elbow also must be exposed if there is a type-2 or 3
coronoid process
fracture11 or
evidence of medial collateral ligament instability. A full-thickness
fasciocutaneous medial flap is elevated, and the ulnar nerve is identified.
The cubital retinaculum is released to allow mobilization of the ulnar nerve.
A tape is placed around the nerve, but it is not clamped so that an assistant
cannot inadvertently pull on the nerve. The flexor carpi ulnaris is released
from the subcutaneous border of the ulna by dividing the deep fascia
approximately 2 mm from its insertion into the ulna. Elevation of the flexor
carpi ulnaris muscle provides exposure to the coronoid process and the medial
collateral ligament (Figs. 6-A
and 6-B). The medial
collateral ligament is usually disrupted from its insertion onto the medial
epicondyle and is simply repaired with transosseous sutures or suture
anchors11.
With a coronoid process fracture, the anterior capsule remains attached to
the coronoid process fragment. In a type-III coronoid process fracture, the
medial collateral ligament is attached to the coronoid fragment as well. The
posterior band of the medial collateral ligament is usually avulsed from the
humerus. Fixation of the coronoid process is obtained by passing a number-5
FiberWire suture (AR-7210; Arthrex, Naples, Florida) through the anterior
capsule as it inserts into the coronoid process. The suture is then passed
through the proximal aspect of the ulna to its subcutaneous border, in
preparation for ligation.
Metallic Prosthesis
Care is required to ensure that the appropriate neck length has been
excised to create a stable rim of neck on which the prosthesis will sit. The
correct diameter of the radial head prosthesis is selected by comparing the
excised radial head fragments and the trial prosthesis. The height of the
prosthesis is best selected to ensure that it will have a normal articulation
with the proximal radioulnar joint, so that it is at the same height as the
trochlear notch. This is the best guide for determining the height of the
prosthesis. It should be congruent and have smooth motion with the capitellum
throughout the full range of motion. When viewed on fluoroscopy, the medial
elbow joint space should be congruent and not have an abnormal gap medially or
laterally. To control the radius for insertion of the prosthesis, a nylon tape
is placed around its neck and is wound tightly. When clamped, it provides a
handle to control the position of the proximal part of the radius.
Alternatively, an AO alligator clamp can be used. The medullary canal of the
proximal aspect of the radius is prepared with the provided broaches. The
appropriate-sized radial head prosthesis is then inserted. The stability of
the joint and prosthesis is assessed by taking the elbow through a full range
of motion. If there is any doubt about the stability of the elbow, fluoroscopy
should be used to confirm stability.
Figure 7 demonstrates
relative stability of the elbow, depending upon the medial collateral ligament
competence and the presence of the radial head. In the cadaveric model,
excision of the radial head can be seen to have a minor effect on elbow
stability (position 1). This is because the medial collateral ligament, which
is the primary stabilizer against a valgus force, remains intact. In a patient
who has had enough force applied to the elbow to produce extensive comminution
of the radial head, there is a high likelihood of an associated medial
collateral ligament injury. The diagram demonstrates that, in this situation,
if the patient is managed with only a radial head excision (position 2), the
joint has considerable instability, as has been confirmed clinically.
Insertion of a metallic radial head prosthesis is likely to improve the
stability of the elbow to position 3. It is important to note, however, that
elbow stability is still compromised, as the medial collateral ligament
remains unstable. To optimize the stability of the elbow, the torn medial
collateral ligament should be repaired, improving the stability, ideally back
to position 4, to allow the surgeon to have the confidence to mobilize the
elbow.
Closure
The suture used to stabilize the coronoid process is ligated first, with
the elbow held in 90° of flexion. The medial collateral ligament complex
is ligated with the elbow held in 30° of flexion and the forearm in full
supination. The lateral collateral ligament complex is stabilized with
transosseous interrupted sutures, as described by Osborne and
Cotterill15, with
the elbow held in 30° of flexion and the forearm in full pronation
(Fig. 8).
Postoperatively, the patient is managed with a dynamic elbow suspension
splint (Fig. 9), which is
created by wrapping a pediatric knee immobilizer around the extended elbow.
The arm is then suspended with a pulley and counterweight of 2 to 4 kg from a
Balkan frame. The weight is attached to the midposition of the upper
extremity, so that it lifts the entire limb and maintains the elbow in
extension. Suspending the arm from the wrist pulls the elbow into flexion and
should be avoided. The dynamic elbow suspension splint rests the extended
elbow in an elevated position, substantially reducing elbow edema, minimizing
the risk of creating a fixed flexion deformity, and taking pressure off the
wound on the posterior aspect of the elbow while it is healing. On the second
postoperative day, the arm is removed from the splint to allow active and
passive mobilization. The patient can then be discharged, and the splint is
set up at home and used for an additional
week16. Patients
are prescribed indomethacin for a period of six weeks to minimize the risk of
heterotopic bone formation.