Supracondylar humeral fractures are the most common elbow fractures in
children1,2.
These fractures are classified, according to Gartland's criteria, as
nondisplaced fractures (type I), hinged fractures with the posterior cortex
intact (type II), and completely displaced fractures (type
III)3. Completely
displaced (type-III) fractures may be associated with neurovascular injuries,
and treatment may be complicated by malunion, elbow stiffness, iatrogenic
neurovascular injury, and compartment
syndrome1,2,4.
The standard treatment for completely displaced (type-III) extension
supracondylar fractures of the humerus in children is closed reduction and
percutaneous pin
fixation1,2.
However, controversy persists regarding whether lateral or crossed medial and
lateral pin fixation is the optimal technique. For the sake of
comprehensiveness, when discussing pin fixation techniques for supracondylar
fractures, we vprefer to describe both the location of the entry of the pins
and the location of the pins in the distal part of the humerus. Thus, pin
fixation may be performed with lateral entry or medial and lateral entry, and
the pins may then be located in the lateral column, the central column, or the
medial column of the distal part of the humerus. Ideally, medial and lateral
entry pins engage the medial and lateral columns at the fracture site, whereas
lateral entry pins engage the lateral and central columns at the fracture
site.
The advantage of medial and lateral entry pin fixation is probably
increased biomechanical stability, although iatrogenic ulnar nerve injury may
result from placement of the medial
pin5-7.
Conversely, the advantage of lateral entry pin fixation is avoidance of
iatrogenic ulnar nerve injury, although the construct may be less stable
biomechanically8-12.
The purpose of this study was to compare the efficacy of lateral entry pin
fixation with that of medial and lateral entry pin fixation of completely
displaced (type-III) extension supracondylar fractures of the humerus in
children. The null hypothesis was that there was no difference between the pin
fixation techniques in terms of major loss of reduction or iatrogenic ulnar
nerve injury.
The study design was a single-center, prospective, randomized clinical
trial. Approval was obtained from our institutional review board, and informed
consent was provided by all of the patients in the study. Patients were
enrolled from May 2003 to January 2005 at a major tertiary care children's
hospital. Fractures were treated by ten different full-time pediatric
orthopaedic surgeons. All of them had additional post-residency pediatric
orthopaedic surgery training and had treated displaced supracondylar humeral
fractures with both lateral entry and medial and lateral entry pin-fixation
techniques.
The inclusion criteria were an age of three to ten years old and treatment
of a completely displaced (type-III) extension supracondylar fracture of the
humerus within forty-eight hours after the injury. The exclusion criteria were
an age of less than three years old or greater than ten years old, an open
fracture, a fracture requiring open reduction, a fracture requiring
neurovascular exploration, a floating elbow injury, bilateral supracondylar
humeral fracture, a previous ipsilateral elbow fracture, and an inability to
perform a preoperative neurovascular examination.
The patients were randomized to be treated either with lateral entry pin
fixation or with medial and lateral entry pin fixation according to an
assignment, produced by a random-number generator, in a sealed envelope. A
permuted block (block size, 4) randomization design was used. The envelopes
were opened in the operating room, after closed reduction and at the time of
percutaneous pin fixation. Pre-hoc sample-size calculation (nQuery Advisor
6.0, Statistical Solutions, Saugus, Massachusetts) showed that the study had
sufficient power (a = 0.05, ß = 0.20) to detect a 10% difference in
the rates of major loss of reduction between groups. A major loss of reduction
was defined as a change in the Baumann angle of >12° between the
initial postoperative and three-month follow-up radiographs.
Surgical techniques were standardized in terms of the pin location, the pin
size, the incision and position of the elbow used for medial pin placement,
and the postoperative course. General anesthesia was used for all patients,
and all were in a supine position on the operating table. Reductions were
performed under the guidance of fluoroscopy. Children who weighed =20 kg
were treated with 0.062-in (1.6-mm) Kirschner wires, and those who weighed
>20 kg were treated with 5/64-in (2.0-mm) Steinmann pins.
For the lateral entry technique, two pins were inserted from the lateral
aspect of the elbow across the lateral cortex to engage the medial cortex with
the elbow in hyperflexion. The pins could be placed in a parallel or divergent
manner (Figs. 1-A and 1-B). For
the pin construct to be considered acceptable, one pin had to be placed in the
lateral column of the distal part of the humerus and the other had to be
placed in the central column (through the olecranon fossa) of the distal part
of the humerus (Figs. 1-A and
1-B). For the medial and lateral entry technique, one pin was
inserted from the lateral aspect of the elbow across the lateral cortex to
engage the medial cortex with the elbow in hyperflexion. The elbow was then
extended to less than a 90° position to avoid injury to an anteriorly
subluxating ulnar nerve. A small medial incision of 1.5 to 3.0 cm was made
over the medial epicondyle. Superficial dissection was performed to ensure
that the pin was placed in the medial epicondyle and that the ulnar nerve was
not subluxated anteriorly over the medial epicondyle. The medial pin was then
placed, starting in the medial epicondyle and engaging the lateral cortex,
with the elbow extended to <90° and with retraction of soft tissue from
the medial epicondyle. The pin construct was considered to be acceptable if
one pin had been placed in the lateral column and one pin had been placed in
the medial or central column of the distal part of the humerus
(Fig. 2). The pins were bent
outside the skin, and a bivalved long arm cast was applied with approximately
70° to 90° of elbow flexion and neutral forearm rotation.
All patients returned for both clinical and radiographic evaluations at one
week, three to four weeks, and three months. The cast and pins were removed in
the clinic at the three to four-week follow-up appointment.
Clinical evaluation was performed by attending pediatric orthopaedic
surgeons. The surgeons and the patients were not blinded to the type of pin
construct. Clinical evaluation included assessment of the carrying angle,
measurement of the passive range of elbow motion, neurologic and vascular
examination of the extremity, and determination of any complications such as
superficial infection, deep infection, and the need for a reoperation. The
clinical results were graded according to the criteria of Flynn et al., which
are based on the carrying angle and elbow
motion13.
Neurovascular examination was performed preoperatively and at the one-week
follow-up visit. Complete clinical evaluation was performed at the three to
four-week and three-month follow-up visits.
Radiographic evaluation included an anteroposterior radiograph of the
distal part of the humerus and a lateral radiograph of the elbow. Radiographs
were made intraoperatively after pin fixation and at the three to four-week
and three-month follow-up visits. The Baumann angle was calculated on the
anteroposterior radiograph with the method of Williamson et
al.14 at the
three-month follow-up examination, and the change in the Baumann angle between
the intraoperative radiograph made after pin fixation and the three-month
postoperative radiograph was recorded as well. The humerocapitellar angle was
calculated on the lateral radiograph with use of the method of Topping et
al.15 at the
three-month follow-up examination, and the change in the humerocapitellar
angle between the intraoperative radiograph made after pin fixation and the
three-month postoperative radiograph was also recorded. All radiographic
angles were calculated by a pediatric orthopaedic fellow who was not blinded
to the type of pin construct that had been used.
At the three-month follow-up visit, in addition to the clinical and
radiographic evaluations, the parents were asked whether the child had
returned to full function, had minor limitations of function, or had major
limitations of function.
Loss of reduction was determined on the basis of the change in the Baumann
angle. No, mild, and major displacement were operationally defined according
to the criteria reported by Skaggs et
al.16, which were
based on the finding that the Baumann angle varies 6° for every 10° of
humeral rotation on the anteroposterior
radiograph17. No
displacement was defined as a change in the Baumann angle of <6°; mild
displacement, as a change of 6° to 12°; and major displacement, as a
change of >12°.
Iatrogenic ulnar nerve injury was determined by clinical evaluation and was
operationally defined as a postoperative ulnar nerve deficit in a patient who
had had a normal result on the preoperative ulnar nerve examination. The study
protocol dictated that all patients with an iatrogenic ulnar nerve injury
undergo an urgent reoperation to confirm the nature of the injury and to place
the medial pin in a new location.
The Fisher exact test was used to compare categorical data between the two
groups, and the Student t test was used to compare continuous data between the
groups. Statistical analysis was performed with SPSS software (version 14.0;
SPSS, Chicago, Illinois) and SAS software (version 9; SAS, Cary, North
Carolina).
During the accrual period, 217 children were treated for a completely
displaced (type-III) extension supracondylar fracture of the humerus. Of these
217 patients, 153 (71%) met the inclusion criteria. Of these 153 eligible
patients, sixty-six (43%) agreed to participate in the study. Of the sixty-six
patients, five were excluded from the study because they had undergone open
reduction; four, because the surgeon had added pins to the randomized pin
configuration; two, because the fracture was deemed to be type II on review;
two, because the pin configuration was not acceptable according to the
protocol; and one, because the preoperative neurovascular examination was
inadequate. Thus, the study population consisted of fifty-two patients.
The lateral pin-entry group comprised twenty-eight patients. The mean age
was 6.1 years old (range, 3.6 to 8.1 years old). Eighteen patients (64%) were
female. The displacement was posterolateral in ten of the twenty-eight
patients, posteromedial in eleven, and straight posterior in seven. Five
patients had comminution; two, a pulseless viable hand; three, an associated
median nerve injury; and two, an associated radial nerve injury.
The group treated with medial and lateral entry comprised twenty-four
children with a mean age of 5.7 years (range, 3.5 to 8.0 years). Thirteen
patients (54%) were male. Displacement was posterolateral in eleven of the
twenty-four patients, posteromedial in nine, and straight posterior in four.
Seven patients had comminution; two, a pulseless viable hand; four, an
associated median nerve injury; and one, an associated radial nerve injury.
There were no significant differences (p > 0.05) between groups with regard
to any of these variables (Table
I).
No patient in either group had a major loss of reduction
(Table I). Six of the
twenty-eight patients treated with lateral entry and one of the twenty-four
treated with medial and lateral entry had a mild loss of reduction; this was
not a significant difference (p = 0.107).
There were no cases of iatrogenic ulnar nerve injury in either group
(Table I). All median and
radial nerve palsies were present preoperatively, were associated with the
fracture, and resolved spontaneously.
There were no significant differences (p > 0.05) between the groups
regarding the Baumann angle, change in the Baumann angle, humerocapitellar
angle, change in the humerocapitellar angle, carrying angle, elbow extension,
elbow flexion, total elbow motion, Flynn grade, superficial infection, need
for a reoperation, or functional return
(Table I).
Intent-to-treat analysis was performed for the four patients who were
enrolled in the study but then excluded because the surgeon had used
additional pins. Three of these patients had been randomized to the lateral
entry group, and a third, medial pin had been added. The fourth patient had
been randomized to treatment with medial and lateral entry pin fixation, and a
second lateral pin had been added. If these patients were considered to have
had a major loss of reduction, there would still be no significant difference
between the two groups in terms of a major loss of reduction (three of
thirty-one in the lateral entry group compared with one of twenty-five in the
medial and lateral entry group, p = 0.620).
Controversy persists regarding the optimal pin-fixation technique for
completely displaced (type-III) extension supracondylar fractures of the
humerus in children, with the two primary treatment methods involving use of
either two lateral pins or crossed medial and lateral pins. In this
sufficiently powered randomized clinical trial, we found no significant
difference between these two pin-fixation techniques in terms of loss of
reduction or iatrogenic ulnar nerve injury.
The advantage of medial and lateral entry pin fixation is probably greater
fracture stability, although iatrogenic ulnar nerve injury may result from
placement of the medial pin. Conversely, the advantage of lateral entry pin
fixation is avoidance of iatrogenic ulnar nerve injury, although the construct
may be less stable biomechanically.
Biomechanical studies of adul t cadavers and synthetic pediatric bone
models have suggested that medial and lateral entry pin fixation provides
greater torsional rigidity than does lateral entry pin
fixation9,10.
However, in addition to the location of the entry of the pins, the overall
strength of these constructs is related to the divergence of the pins in
different columns of the distal part of the humerus and to the overall number
of pins9. Three
lateral entry pins or two lateral entry pins that are divergent and are
located in both the lateral and the central column provide torsional rigidity
that is similar to that achieved with the combination of a medial and a
lateral pin9.
The reported risk of iatrogenic ulnar nerve injury from medial entry pin
fixation has varied widely and probably depends on the specific technique of
pin insertion. In a recent quantitative synthesis that pooled data from 1680
patients from thirty-three studies that met a priori eligibility criteria, the
rate of iatrogenic ulnar nerve injury from medial and lateral entry pin
fixation was
3.3%18. The
relative risk of iatrogenic ulnar nerve injury was 4.86 times higher for
medial and lateral entry pin fixation than it was for lateral entry pin
fixation. The risk of iatrogenic ulnar nerve injury can be reduced with a
medial incision and with extension of the elbow during medial pin
placement5,19-21.
The majority of iatrogenic ulnar nerve injuries associated with medial pin
fixation resolve after wound exploration and replacement of the medial pin at
a new
location7,22,23.
The reported risk of loss of reduction following lateral entry pin fixation
has also varied widely. Davis et al. reported displacement of two of seven
type-III fractures treated with two lateral
pins12. Kallio et
al. found a loss of reduction in 14% (eleven) of eighty cases in which two
lateral pins had been
used11. However,
Skaggs et al. reported no loss of reduction after fixation of fifty-five
type-III fractures with two or three lateral entry
pins16. In a recent
quantitative synthesis that pooled data from 1680 patients from thirty-three
studies that met a priori eligibility criteria, the rate of displacement
following lateral entry pin fixation was
2.1%18. The risk of
displacement after lateral entry pin fixation can be reduced by emphasizing
proper pin-placement technique, with divergent pins, pins that engage the
lateral and central columns, and use of a third lateral pin if
needed11,16.
Lateral entry pin fixation has been compared with medial and lateral entry
fixation in numerous retrospective case series of extension supracondylar
fractures of the humerus in children. In general, these studies have shown a
risk of iatrogenic ulnar nerve injury in association with medial and lateral
entry pin fixation and a risk of loss of reduction after lateral entry pin
fixation. However, limitations of these studies include potential selection
bias in terms of the pinning technique and the lack of complete data due to
the retrospective study design. In a review of forty-nine type-II fractures
and eighty-nine type-III fractures treated with three different pin fixation
techniques, Gordon et al. found one case of rotational loss of reduction after
treatment of a type-III fracture with two lateral
pins8. In a series
of 141 type-II fractures and 204 type-III fractures, Skaggs et al. reported no
difference in maintenance of reduction between the two methods, but iatrogenic
ulnar nerve injury was seen in 10.6% (seventeen) of 160 cases treated with a
medial pin5. In a
study of forty-seven children in whom a type-III supracondylar fracture had
been treated with crossed pins (twenty-seven patients) or with lateral pins
only (twenty), Topping et al. found no loss of reduction in either group and
one ulnar nerve injury in the group with crossed
pins15. Similarly,
in a study of fifty-six fractures, Shamsuddin et al. found three iatrogenic
ulnar nerve injuries associated with medial and lateral entry pin fixation and
two iatrogenic radial or anterior interosseous nerve injuries associated with
lateral entry pin fixation, although there was no difference in loss of
reduction24.
Foead et al. performed a randomized clinical trial in which thirty-four
type-II or III fractures were treated with medial and lateral pin fixation and
thirty-two were treated with lateral pin
fixation25.
Straight lateral skin traction was used prior to closed reduction and pinning.
Details were not provided regarding whether a medial incision was used for
insertion of the medial pin, the position of the elbow during medial pin
placement, or the pin location with respect to the columns of the distal part
of the humerus. There were no significant differences in terms of loss of
reduction, the Baumann angle, or elbow motion between the two groups. There
were five iatrogenic ulnar nerve injuries in the medial and lateral entry
group, and there were two iatrogenic ulnar nerve injuries and one iatrogenic
radial nerve injury in the lateral entry group.
One of the strengths of the current study was that it was a randomized
clinical trial with the patients randomized as close as possible to the time
of fracture treatment. Also, both the lateral entry and the medial and lateral
entry techniques were standardized in terms of pin size, pin location, the use
of a medial incision, and the position of the elbow for medial pin placement.
Full clinical and radiographic evaluation was performed at standardized
intervals. Weaknesses of this study include the lack of blinding of the
clinical and radiographic observers to the pinning technique. The study had
adequate power to detect a 10% difference in the rate of major loss of
reduction between the techniques. A 10% difference was chosen because it was
the level of difference that would compel the group of pediatric orthopaedic
surgeons involved in this study to change their preferred pin-fixation
technique. Thus, this was thought to be the clinically relevant difference.
Also, the power of the study to detect loss of reduction was determined pre
hoc. There was no significant difference in the rates of iatrogenic ulnar
nerve injury between the groups, but the study may not have had sufficient
power for this end point, and post-hoc power analysis for the end point of
iatrogenic ulnar nerve injury was not performed as this technique is largely
considered to be invalid.
In conclusion, both pin-fixation techniques appear to be effective.
However, these results are contingent on the specific surgical techniques used
in this study. For medial and lateral entry pin fixation, the lateral pin was
placed first, the elbow was extended to a position of <90°, and a small
incision was made over the medial epicondyle to protect the ulnar nerve. Other
techniques, such as blind placement of the medial pin without a small incision
or placing the pin with the elbow flexed >90°, were not assessed in
this study and may be associated with a higher risk of iatrogenic ulnar nerve
injury. For effective lateral entry pin fixation, the pins must be placed in
both the lateral and the central column of the distal part of the humerus.
Other fixation techniques, such as placement of both pins in only the lateral
column, were not assessed in this study and may be associated with a higher
risk of major loss of reduction. Persistent instability occurring after the
placement of two lateral entry pins or crossed medial and lateral entry pins
also was not addressed in this study, as patients with instability after
placement of pins were excluded. The addition of a third pin is usually
recommended in these patients to achieve fracture stability. ?