Skeletal Deformity Pelvis X-Fix
SKELETAL DEFORMITY FOLLOWING EXTERNAL FIXATION OF THE PELVIS
Kyle F. Dickson, MD* and Joel M. Matta, MD**
*Tulane University School of Medicine, Department of Orthopaedic Surgery, New Orleans, LA
**The University of Southern California School of Medicine, Department of Orthopaedic
Surgery, and the Hospital of the Good Samaritan, Los Angeles, CA
Address Correspondence to:
Kyle F. Dickson, MD
Tulane University School of Medicine
Department of Orthopaedic Surgery SL32
1430 Tulane Avenue
New Orleans, Louisiana 70112-2699
(504) 584-3515 / Fax: (504) 584-3517
Presented in part at the Western Orthopedic Association in Santa Barbara, 1995 and AAOS
Annual Meeting, March 19, 1998, New Orleans, Louisiana
Skeletal Deformity Pelvis X-Fix
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
SKELETAL DEFORMITY FOLLOWING EXTERNAL FIXATION OF THE PELVIS
ABSTRACT
OBJECTIVE: To better define the three-dimensional deformity of acutely injured unstable pelvises
prior to and after application of an external fixator emergently using recently defined measurement
techniques.
DESIGN: Retrospective
SETTING: Large pelvis fracture referral practice
PATIENTS: Sixteen out of 151 patients referred to our institution who were hemodynamically
unstable and had a mechanically unstable pelvic injury after emergent application of an external
fixator by the referring orthopaedist prior to transfer.
MAIN OUTCOME MEASUREMENTS: We reviewed the anterior posterior, inlet, outlet, and CT
radiographs both prior to and after placement of the external fixator. A 3-dimension X, Y, and Z
axis was used to define the deformity (No figure citations in an Abstract). Each axis defines a
rotational deformity and a translation deformity (i.e. x-axis – flexion/extension and
diastasis/impaction, y-axis – internal/external rotation and cephalad/caudad translation, z-axis –
abduction/adduction and anterior/posterior translation). (You don’t say anything about measurements
for the translational and rotational deformities as “Main Outcome Measurements”)
RESULTS: The most common deformities found were flexion (7/16), internal rotation (8/16),
posterior translation (7/16), and cephalad translation (12/16). Sixty-seven percent (8/12) of the
patients had worsening of the deformity posteriorly despite apparent improvement anteriorly on the
AP radiograph. Although many deformities existed, we defined external fixator deformity as
flexion and internal rotation of the hemipelvis, as this deformity it consistently worsened with
an anterior frame. Seventy-three percent (8/11) of the patients had worsening of the external
fixation deformity. Sixty-seven percent (8/12) patients had worsening of the posterior translation or
diastasis. After placement of the external fixator, all patients displayed greater than 1 cm of either
posterior cephalad translation or diastasis (average 3.4 cm range 1.3 cm to 4.6cm).
CONCLUSION: An external fixator deformity may occur (flexed internally rotated hemipelvis).due
to the forces placed on the pelvis during reduction by the surgeon and stabilization by an external
fixator. Furthermore, 67% of the patients had an increase in posterior displacement with placement
of an external fixator. Understanding the three-dimensional deformity of the hemipelvis after
Skeletal Deformity Pelvis X-Fix
2
37
38
39
40
41
42
43
44
45
46
47
injuryenables the surgeon to more easily perform the proper reduction maneuvers necessary
to obtain a more anatomical reduction using external fixation.
48
The acute management of pelvic trauma is controversial. Possible initial stabilization treatments for
49
acute pelvic injuries include no external fixators11 (medical stabilization with possible
50
angiography), anterior fixators 19, and posterior fixators.6,9,29 Numerous authors support the use of
51
an external fixator in the acute hemodynamically unstable patient with a mechanically unstable
52
pelvis.4,7,8,10,12,13,15,19,23,24,28
53
population. Some clinical2,11,20,21,27, biomechanical14, and anatomical3 evidence has questioned the
54
use of external fixators. Furthermore, the use of external fixators alone for the definitive treatment
55
of Bucholz type III3 and Tile type C26 unstable pelvic injuries has resulted in unacceptable rates of
56
failure.14, 18 Most important, associated injuries and coagulopathies contribute more to hemodynamic
57
instability than the pelvic injury.3,20
KEY WORDS: Pelvis Fractures, external fixation, Pelvis Deformity, Unstable Pelvis, Measuring
Techniques
INTRODUCTION
However, all of these studies have major flaws in their control
58
59
Despite the controversy on the use of anterior external fixators, many authors recommend its use to
60
reduce and stabilize a hemodynamically unstable patient with a mechanically unstable pelvis in the
61
acute setting. There have been several authors who have previously described the deformity of the
62
pelvis however, these studies reported the displacement of the pelvis with a linear
63
measurement.23,24,28 No study has reported on the actual three-dimensional deformity of the pelvis,
Skeletal Deformity Pelvis X-Fix
3
64
and what is more important, the three-dimensional deformity of the pelvis after external
65
fixation.2,21,27 Due to the increasing number of pelvic malunion referrals that were seen at our
66
institution, we analyzed the acute deformity of the pelvis both prior to and after placement of an
67
external fixator.17 Our purpose is to further define the three-dimensional deformity of the acutely
68
injured pelvis prior to and after application of an external fixator emergently using recently defined
69
measurement techniques.
70
incorrectly in the emergent life saving conditions might potentially place the vascular
71
injury/clot at further risk for continued or worsening hemorrhage.
72
(Is it correct or fair to say, that you were never able in this study to scientifically determine that “The
73
hypothesis was that the anterior external fixator placed in the emergent life saving conditions
74
incorrectly might potentially place the vascular injury/clot at further risk for continued or worsening
75
hemorrhage”? If you agree with that premise, then there is no need to use your statement as the
76
hypothesis for your study. Your hypothesis could be something like what the conclusion is in your
77
abstract, “Understanding the three-dimensional deformity of the hemipelvis after injury enables the
78
surgeon to more easily perform the proper reduction maneuvers necessary to obtain a more
79
anatomical reduction using external fixation”.
80
81
82
The hypothesis was that the anterior external fixator placed
Skeletal Deformity Pelvis X-Fix
4
83
MATERIALS AND METHODS
84
One hundred and fifty-one pelvic fractures were referred to our institution for definitive
85
operative treatment (1986-1994) (Table 1). Sixteen patients were considered hemodynamically
86
unstable (systolic blood pressure less than 90 mm Hg) and had an external fixator placed emergently
87
by the referring orthopaedist prior to transfer. Of the sixteen patients, fourteen were unstable
88
Bucholz type III and two were Bucholz type II injuries .3 According to the Tile Classification, there
89
were four C3, one C2, nine C1, and two B1. We reviewed the anterior posterior view (AP), caudad
90
(inlet), cephalad (outlet), and CT scans prior to and after placement of the external fixator on each of
91
the sixteen patients. No pre-external fixation films were available in three cases (In Table 1, you
92
list four cases of “No pre op x-ray” or “unusable pre op x-ray” [Cases #1, 11, 12, 13]) . However,
93
the post ex-fix radiographs were used to define the deformity. Deformities analyzed include leg
94
length discrepancy, flexion/extension, internal/external rotation, abduction/adduction, and translation
95
of the hemipelvis (Table 2). Only the CT scan can be used reliably to measure deformity,
96
especially rotational deformities. Plain x-rays define the type of deformities, and specific
97
measurements only for r, pure translational deformities (Figure 4).
98
99
Prior to defining the deformities of the pelvis, one must understand that the deformity exists in three-
100
dimensional space. Many of the deformities cannot currently be measured with an exact number of
101
degrees. However, by using various radiographic landmarks and their relationships to each other,
102
deformities of the pelvis can be defined. For example, although the surgeon cannot measure the
103
exact degree of abduction, he or she can measure symphyseal diastasis and determine that the amount
104
of abduction has increased if other deformities remain unchanged. Because all deformities are
Skeletal Deformity Pelvis X-Fix
5
105
interrelated, the surgeon has to ensure this diastasis does not result from an additional
106
translation. We attempt to define the deformity by a 3 axis (x, y, and z) system positioned in
107
three-dimensional space, (Figure 1). Similar to a long bone, once a 3-D CT program becomes
108
available, a resultant vector in space can be used to define the pelvic deformity. Due to the
109
three-dimensional nature of the deformity, all the individual deformities are interrelated. For
110
example, elevation of the superior rami may be a combination of either flexion or cephalad
111
translation. Comparing the posterior translation will confirm the deformity (no posterior cephalad
112
translation means the elevation of the rami is flexion because flexion is based in the posterior
113
pelvis).19 (What would be the converse of that statement, if one exists? Posterior cephalad
114
translation means the elevation of the rami is not in flexion because posterior cephalad translation is
115
based ???.) The displacement of the hemipelvis in space is quite complex and in most cases cannot
116
be defined in actual degrees or millimeters. However, the deformity of the pelvis can be broken
117
down into three rotational and 3 translational deformities that can allow the orthopaedic surgeon a
118
better understanding of the reduction vector required to reduce the hemipelvis. Furthermore,
119
although actual measurements cannot be compared, viewing the pre and post external fixator, AP,
120
inlet and outlet views, the surgeon can easily see which particular deformities improved or worsened
121
with external fixator application. Although we recognize the displaced pelvis does not rotate around
122
a single point supra-acetabular, we are developing a 3-D pelvic program that can define the
123
normally positioned pelvis and use a three coordinate axis to define a resultant vector of the
124
displaced hemipelvis. The anatomically placed hemipelvis in space aligns the anterior superior iliac
125
spine and the pubis symphysis in the frontal plane and the pubis symphysis and the center of the
126
sacrum on the sagittal plane. A comparison to this anatomically placed hemipelvis will define the
Skeletal Deformity Pelvis X-Fix
6
127
deformity. This enhances our understanding of the actual three-dimensional deformity of the pelvis.
128
(Did you use any of the 3D pelvic programs in the preparation of this manuscript or is it too
129
preliminary to do so? Second, why do you mention the program in this section rather than in the
130
discussion where it fits better as to future plans to study pelvic deformity?)
131
132
Our definition of these various deformities is as follows.
Translation of the pelvis is
133
anterior/posterior (z-axis, Figure 1) or cephalad/caudad (y-axis, Figure 1).
134
diastasis/impaction is a translation on the x-axis (Figure 1). Measuring cephalad/caudad translation
135
on the AP view is easily performed (Figure 4). Classically, posterior displacement is defined on the
136
inlet view. However, direct cephalad translation of the hemipelvis will cause an apparent posterior
137
translation on the inlet view. Therefore, the posterior translation is best measured on the CT scan.
138
The actual cephalad translation is estimated on the AP view from a line in the plane of the sacrum
139
(Figure 4). A perpendicular distance from this line to the ischium and the acetabular dome
140
demonstrates the amount of cephalad translation. The distance is compared to the contralateral
141
hemipelvis. A difference in ischial heights creates a sitting imbalance. A difference in acetabular
142
dome heights creates a leg length deformity.5,17 The cephalad translation distances are an
143
estimation. Flexion/ extension and abduction/adduction of the hemipelvis can alter this translation
144
measurement. (Somewhere in the beginning, perhaps in the Abstract or better in this section, it can
145
be summarized there are three deformities on three axes that can be accurately measured on plain x-
146
rays as pure translational deformities (anterior/posterior, cephalad/caudad and diastasis/impaction).
147
[Correct statement?] There are also three rotational deformities that are measured on three axes that
148
are better defined and measured by a CT scan (flexion/extension, internal/external rotation, and
Furthermore,
Skeletal Deformity Pelvis X-Fix
7
149
abduction/adduction). [Is this a correct statement?]. Other deformities that be indirectly measured
150
include leg length discrepancy and sitting imbalance and ???)
151
The rotational deformities include flexion/extension (x-axis), internal/external rotation (y-axis), and
152
abduction/adduction (z-axis). Flexion/extension of the hemipelvis is defined as the rotation of the
153
hemipelvis around an axis that passes supra-acetabular from lateral to medial through the iliac wing
154
(x-axis, Figure 1). In reality, most of the flexion/extension is based in the posterior pelvis
155
(reference?).
156
hemipelvis. They are: (1) the obturator acetabular line16 to the teardrop (the more cephalad the line
157
crosses the teardrop, the more flexion of the hemipelvis), (2) the shape of the obturator foramen on
158
the outlet or AP view (the foramen becomes more elongated and elliptical with flexion), and (3) the
159
position of the ischial spine within the obturator foramen on the outlet view (the more caudad the
160
ischial spine is in relationship to the foramen the more flexion). The best actual measurement of
161
flexion is obtained from the three-dimensional CT scan. The normal hemipelvis and sacrum are
162
removed from the anatomically positioned pelvis. An angle is measured from a line drawn between
163
the anterior superior iliac spine (ASIS) to the symphysis and a line perpendicular to the floor.
164
Unfortunately, no pre-external fixator 3-D CT’s were available to measure an actual degree of
165
flexion. However, when comparing the relationships listed above, we can define whether a
166
flexion deformity worsened, stayed the same, or improved with external fixation application.
Radiographic relationships are also used to define flexion/extension of the
167
168
Internal/external rotation of the hemipelvis is defined around an axis that is perpendicular to the floor
169
and splits the quadrilateral surface (y-axis, Figure 1 & 5A). Defining internal rotation on plain films
170
is performed by (1) comparison of widths between the two ischia (increased width with internal
Skeletal Deformity Pelvis X-Fix
8
171
rotation), (2) width of an iliac wing (greater with external rotation), and (3) the relationship of the
172
ilioischial line to the teardrop (the more lateral the line the more internal rotation). The CT scan
173
more precisely defines the degree of rotation. Drawing a line parallel to the very constant
174
quadrilateral surface (two to five mm above the dome) and the angle, this forms with a horizontal
175
line in the plane of the sacrum measures rotation solely (Figure 2a-f, 3). Sponseller25 used the line
176
from the ASIS to the posterior superior iliac spine (PSIS) to measure deformity of the hemipelvis in
177
children with congenital pelvic deformity. However, this measurement is a combination of
178
internal/external rotation and abduction/adduction rotation.
179
180
Abduction/adduction deformity is defined as the rotation of the hemipelvis around an axis that is
181
perpendicular to the axis of internal and external rotation. This axis passes anterior posterior through
182
the supra-acetabular region (z-axis, Figure 1 & 5B). Pure abduction/adduction will not affect the
183
internal/external rotation measurement. Pure abduction/adduction deformities are rare and usually
184
associated with other rotational deformities. One can determine whether an abduction/adduction
185
deformity exists on the inlet view if no other rotational deformity exists. The angle formed by a line
186
from the posterior superior iliac spine (PSIS) to the symphysis pubis and a line in the plane of the
187
sacrum estimates the abduction/adduction deformity.
188
abduction/adduction can be determined from the CT scan. After ruling out diastasis or impaction
189
(lateral translation), comparing the distance from the center of the quadrilateral surface to the midline
190
of the injured side to that of the non-injured side estimates the relative amount of
191
abduction/adduction deformity but does not give the actual degree of rotation. (Figure 3). Pure
An estimation of the amount of
Skeletal Deformity Pelvis X-Fix
9
192
abduction and adduction are rare and usually associated with external rotation and internal
193
rotation deformity (Figure 5). (Lines 172-173 say the same thing. Use this statement just once)
194
195
Displacement of the hemipelvis represents a combination of rotational and/or translation
196
deformities (i.e. cephalad translation of the ischium could be a pure cephalad translation or a
197
combination of rotational (flexion) and translational (cephalad) deformity). Pre and post
198
external fixation films were reviewed to determine the rotational deformities (i.e. flexion/extension,
199
internal/external rotation, and abduction/adduction) and whether these rotational deformities
200
worsened with the external fixator. Cephalad displacement of the posterior and anterior aspect
201
of the pelvis were performed using the AP radiograph and diastasis of the fracture or
202
sacroiliac joint were measured by the CT. (Figure 3)
203
204
Most deformities cannot be quantified until improved software for the 3-D CT is available, i.e.
205
flexion, extension.
206
deformation, i.e., flexion/extension of the hemipelvis was measured on the AP and outlet but
207
not on the inlet view. Furthermore, internal rotation was marked as either present or absent
208
depending on the radiographic landmarks mentioned above. The magnitudes of the linear
209
measurements were used to compare pre external fixator views to post external views.
210
However, the linear measurements could not be translated into the actual degree of rotation of
211
the deformity. Therefore, the magnitude of the measurements was used for comparison only.
212
The surgeon must view the whole series of radiographs as a unit to define the deformity. For
213
example, , on the AP film, an elevation of the rami at the hemipelvis may represent either
Certain views on plain radiographs were used to qualify specific
Skeletal Deformity Pelvis X-Fix
10
214
translation or flexion of the hemipelvis; a comparison then must be made using the outlet view
215
and the cephalad translation of the posterior aspect of the pelvis. With regard to the results as
216
in Table 1: Questions and summations. (Please see “Legend or Table 1 page as well)
217
1. In the “Flexion” and “Internal Rotation” columns, can AC FX be used instead of just “AC”?
218
2. In the “Final Reduction After Internal Fixation” column, what does the “3” in “3-STAGE”
219
stand for and what does stage(d) pelvic reconstruction mean in terms of your definitive
220
treatment for those patients versus the other patients who did not have “3-STAGE”?
221
3. In the Figure Legend Page for Table 1, it appears as though your cm definition for reduction
222
post ORIF as to whether poor, good, or excellent is backwards if it is residual centimeters
223
post surgery?
224
4. In the “Loss of Reduction” column, you do make it clear in the text what time period you are
225
talking about; presumably loss of reduction at your institution prior to definitive fixation
226
(the three “yeses” in that column. Please clarify as patients #11, 12, and 13 who had no
227
preoperative or usable injury films did not have a recorded reduction [blank space], yet
228
patient #1 who also had no preoperative or injury film does have a recorded reduction [NO].
229
5. In the column “Greatest Posterior Displacement”, patient #16 has a 1.5 cm diastasis “+H9”.
230
What is “+H9”?
231
6. With regard to summations in various columns in Table 1, the following have been tabulated:
232
A. In the “Greatest Posterior Displacement” column, there were seven cephalad
233
translations and nine diastases. If the cephalad and diastases are lumped together, the
234
range according to Table 1 is from 1.5 cm to 5.4 cm, not the figures give in the
Skeletal Deformity Pelvis X-Fix
11
235
results (1.3-4.6 cm). Furthermore the average is 2.93 cm, not the 2.4 cm listed in the
236
results.
237
B. In the “Loss of Reduction” column, there are 10 “nos” and 3 “yeses”. The three
238
blanks are in patients #11,12, and 13. Patient #1 as mentioned above who also had no
239
preoperative x-ray, has a recorded “no”. In your text you state “Four out of sixteen
240
patients did not have enough comparable films pre external fixation to judge the
241
adequacy of the external fixator”. Please clarify the status of patient #1 regarding this
242
statement and inclusion in the column as “no” loss of reduction while the other three
243
patients have a blank in that column.
244
C. The column labeled “Final Reduction After ORIF” has 7 excellent, 6 goods, one poor
245
and 2 3-stages. You do not appear to mention this information in the text and
246
therefore the question is why is this information in Table 1?
247
D. The column labeled “Flexion” had 6 no, 5 yes, 2 AC and patients #11, 12, and 13 are
248
blank recordings. The exact same numbers are in the next column “Internal
249
Rotation”. What information in the text uses this data?
250
251
E. With regard to “Deformity”, the calculations show 16 patients with the following
deformities and agree with your text as the most common ones:
252
i.
7 abduction
253
ii.
6 adduction
254
iii.
3 extension
255
iv.
7 flexion
256
v.
8 posterior
Skeletal Deformity Pelvis X-Fix
12
257
vi.
13 cephalad
258
vii.
9 internal rotation
259
viii.
2 external rotation
260
There are no recordings for the following deformities; impaction, caudad, anterior or diastasis. Is this
261
because impaction, caudad and anterior deformities were not seen and diastasis is? Please clarify.
262
263
264
Skeletal Deformity Pelvis X-Fix
13
265
RESULTS
266
After placement of the external fixator, all patients had greater than one centimeter of either cephalad
267
translation or diastasis measured in the posterior aspect of the pelvis with the average displacement
268
of 2.4 cm (range 1.3 cm to 4.6 cm). This was due to a combination of the injury and placement
269
of the external fixation. The most common deformities were flexion (7/16), internal rotation
270
(9/16), posterior (8/16) and cephalad (13/16) translation. Abduction (7/16) and adduction (6/16)
271
deformities were almost equally represented. The deformities pre-external fixators were similar
272
to those post external fixator, i.e. if posterior diastasis was present pre-external fixation it was
273
present post external fixation, only worse. (Does that mean that every posterior diastasis
274
worsened after placement of the external fixation (8/16 patients? There is some confusion re the
275
words ”posterior diastasis”. “Diastasis” is listed as a translational deformity along with impaction on
276
the x axis and “posterior” is listed as a translational deformity on the z axis. The terminology
277
“posterior diastasis” appears to combine one term from the translational x and z axes together. Please
278
clarify or use a different term throughout the paper.Worsening is defined as > 2 mm change on any
279
linear measurement. The average internal rotation deformity of our patients equaled 22 (range 8 to
280
52). We were able to compare the rotational deformities, both before and after external fixator
281
placement, for only eleven patients because of associated acetabular injuries, lack of injury
282
radiographs, or both. Surprisingly, six out of eleven (55%) patients had worsening of the flexion
283
deformity and five out of eleven (45%) had worsening of the internal rotation deformity. The
284
"external fixation deformity" is defined as worsening of either flexion, internal rotation, or both.
285
The accident caused the deformity. However, the external fixator made it worse in eight out of
286
eleven (73%) of the cases (5 cases could not be evaluated for deformity due to acetabular
Skeletal Deformity Pelvis X-Fix
14
287
fractures, inadequate preop films, or no flexion internal rotation deformities). (You said the
288
almost the same thing in lines 223-224. Please use statement only in one place).
289
290
The anterior measured change in diastasis of the symphysis after external fixation ranged from 3.8
291
cm improvement to 1 cm worsening with the average being an improvement of 7 mm. Posteriorly
292
the cephalad/caudad translation ranged from 2.4 cm improvement to 1.5 mm worsening for an
293
average of 4 mm worsening. Sacroiliac diastasis ranged from 2.1 cm improvement to worsening of
294
2.7 cm and averaged 2mm of improvement.
295
296
Eight out of twelve patients (67%) had worsening of the deformity posteriorly (either translation
297
cephalad or diastasis) with the placement of an anterior external fixator. Three patients lost
298
reduction in the external fixator prior to definitive treatment with open reduction internal fixation.
299
Two of these patients had worsening at the cephalad translation in the posterior aspect of the pelvis
300
and one had worsening of the symphysis diastasis. (Table 1) The final position after loss of
301
reduction was used for evaluation of deformity. Four out of sixteen patients did not have enough
302
comparable films pre external fixation to judge the adequacy of the external fixator. However, these
303
four could (most probably had?) have the deformity defined by the post external fixation films. (Are
304
you saying for 100% sure that whatever the deformity was on the post external fixation film, that that
305
deformity existed on an injury film (which you don’t have) and did not change except in degree on
306
the post external fixation film? Therefore, because the application of the external fixator did not
307
cause any new deformity, or get rid of a previous deformity, you can use that post external fixation
308
film as if it were the injury film, a surrogate if you will? Or is this too definitive a statement as
Skeletal Deformity Pelvis X-Fix
15
309
theoretically a deformity present on the injury film could have been corrected to anatomic on the post
310
external fixation film and you would not be aware of it as you do not have the injury film?)
311
312
Skeletal Deformity Pelvis X-Fix
16
313
DISCUSSION
314
The use of external fixators alone for the definitive treatment of unstable pelvic injuries has resulted
315
in unacceptable rates of failure.14.18 However, the placement of an anterior external fixation frame in
316
the acute pelvic injury for initial stabilization (for hemodynamic stability?) has been supported by
317
many authors.4, 8,10,12,13,15,17,19,23,24,28 However, some articles question this treatment.11,19 Our study
318
does not preclude the use of external fixation as an acute resuscitation frame. Instead, it highlights a
319
potential problem an orthopedist may encounter when using this treatment. The authors still
320
believe the anterior frame should be placed in life threatening hemodynamically unstable
321
patients with a mechanically unstable pelvis. However, to prevent potential worsening of the
322
reduction posteriorly, the traumatologist should focus on the posterior reduction not the
323
anterior reduction. (But is that absolutely necessary when the referring doctor is only applying the
324
device for temporary use as a hemodynamic resuscitator, knowing that your institution is going to do
325
definitive fixation? Is it their responsibility as the first responder to a hemodynamically unstable
326
patient with a pelvic fracture to use further time to obtain a better reduction with the external fixator,
327
or transfer the patient as quickly as possible? In that scenario, does it then become your responsibility
328
as the final treating physician to ascertain what the deformities are after transfer, and to correct them
329
with definitive fixation, rather than the referring physician? The goal or hypothesis of your paper is
330
“In conclusion, the surgeon placing an external fixator for a completely unstable hemi-pelvis must
331
have a thorough knowledge of the 3-D deformity of the pelvis and the force vectors required for
332
reduction to prevent potential worsening of the deformity and hemodynamic status”. Since there is
333
no hard evidence that the hemodynamic status is made worse by an increasing malreduction post
334
emergent external fixation, what does the referring physician do in the best interest of the patient;
Skeletal Deformity Pelvis X-Fix
17
335
transfer ASAP or study films and spend the time necessary to further reduce the fracture? The
336
question also exists that when you received a patient with an applied external fixator that showed a
337
worsening deformity [and by corollary a potential source for continued or further hemodynamic
338
instability], did you obtain a CT scan and other films as necessary, and emergently re-reduce the
339
fracture prior to your definitive fixation to correct the reduction and/or potentially improve
340
hemodynamic instability? In reality, did you do any external fixator re-reductions prior to definitive
341
fixation to improve the reduction or the reduction that was lost in 3 patients while awaiting definitive
342
therapy? It seems that at least emergently, the rationale for the transferring physician to obtain the
343
best reduction possible is the lessening of hemodynamic instability, which has not been proven in
344
this paper. However, your paper points out well, that if the three dimensional measurements can be
345
obtained quickly and understood in an emergent situation, then that information can be intelligently
346
used by the referring physician to get the best reduction possible before transfer.
347
Understanding both the pelvic deformity and the necessary reduction force gives the surgeon a better
348
chance of preventing an external fixator deformity.23,24 73 % of the patients had worsening of
349
internal rotation or flexion of the hemipelvis causing increased displacement posteriorly. It is likely
350
that the pins of the external fixator in the unstable pelvis were pushed together anteriorly (Is there a
351
better set of words to describe what you mean than other than “the pins of the external fixator in the
352
unstable pelvis were pushed together anteriorly”? Do you mean during frame tightening of the
353
external fixator the pins were translated toward the midline?) during the reduction causing internal
354
rotation, a flexion, and diastasis in the posterior ring of the pelvis. A poorly reduced posterior
355
ring may contribute to more mechanical and hemodynamic instability in a patient.1 The
356
surgeon applying the anterior external fixator must be aware of this deformity. If this deformity
Skeletal Deformity Pelvis X-Fix
18
357
exists, the surgeon can (and should?) obtain a better reduction.using a combination of leg
358
traction and manipulation of the posterior pelvis with compression,.
359
360
The cephalad translation or diastasis worsened in the posterior aspect of the pelvis in eight out of
361
twelve patients (67%) after application of an anterior external fixator frame. This is a worrisome
362
number. The anterior symphysis injury often improved with the fixator. However, the main area of
363
stabilization (define “main area of stabilization”) posteriorly worsened. Furthermore, the reduction
364
posteriorly had greater than one-centimeter displacement (average 2.4 cm, range 1.3 – 4.6 cm). Most
365
of these deformities were greater than 1 cm prior to attempted reduction. (Do you have the range and
366
average of those measurements?) Although beyond of the scope of this paper, the best long-term
367
results after definitive fixation occur when there is less than one centimeter of displacement.22 This
368
failure of reduction may explain the high rate of failure associated with the use of external fixation
369
as the definitive treatment of unstable pelvic injuries.14,18
370
371
Although these external fixators were not applied by experienced trauma orthopaedic
372
surgeons, (With all due respect, how do you know this, and is there a proven difference or
373
consequence, if you as a referral center are going to do “definitive fixation” anyway? Hypothetically,
374
if you were the referring physician applying the emergent external fixator, based on what you now
375
know, would your results in that situation be a significant improvement in the patient’ status on
376
arrival to the trauma center? Tthe high rate of worsening of the posterior ring deformity
377
underscores the need to understand the mechanics of anteriorly placed external fixators for the
378
unstable pelvis with complete posterior disruption (Has this “alarmingly” high rate of worsening
Skeletal Deformity Pelvis X-Fix
19
379
of the pelvic ring deformity made any clinical difference in the outcome of the patients you
380
definitively treated acutely?. If you didn’t change any of these external fixators prior to definitive
381
therapy, especially the three that lost reduction awaiting definitive therapy, your stance re worsening
382
and the need to do something about it has less impact. (The preceding sentence has no scientific
383
meaning and therefore is deleted). Assuming that the more experienced trauma orthopaedist will
384
achieve a better reduction (only when doing definitive fixation, probably not with re-manipulation of
385
an anterior frame), these findings are still important because many community hospitals are using
386
external fixators as initial stabilizers as well as the definitive form of care for unstable pelvic
387
injuries. (Is it better or worse for the patient if the temporary external fixation has been put on as a
388
potential hemodynamic stabilizer and transfer frame, and the transfer doctor knows that the
389
deformities are not corrected and may be worsened by fixator application; yet feels the potential
390
positive effects on attaining hemodynamic stability outweigh the negatives? That, in addition to the
391
fact that the transferring surgeon expects you to definitively correct any deformities by other
392
methods?)
393
Is it also important to be sensitive in the above paragraph not to make distinctions between the
394
referral center and the “community hospital” and a less experienced orthopaedic traumatologist [the
395
“them and us” syndrome]. There are experienced orthopaedic traumatologists in community
396
hospitals who are certainly the equivalent of those in referral centers. It is best not to use statements
397
that do not have any scientific and educational value, even if they may be anecdotally true in some
398
situations (which itself is speculative). Moreover, the purpose of the paper to present scientific
399
information to improve the care of hemodynamically unstable pelvic fracture patients by alerting the
Skeletal Deformity Pelvis X-Fix
20
400
orthopaedic community as a whole to the advantages (real or theoretical or both) of understanding
401
pelvic fracture deformity.
402
A potential error and limitation of this study was the comparison of linear measurements pre
403
and post external fixation. The CT scan measurements were very accurate (posterior diastasis,
404
internal rotation, and abduction). More problematic are the AP films, which could have
405
different magnifications and angles of exposures. This difference could effect the reliability of
406
cephalad translation measurements. Further studies need to be completed on the reproduction
407
of cephalad translation measurements. The authors however, believe these differences in
408
techniques were minimal and the measurements were accurate. (Another potential error or
409
limitation is the assumption that in the absence of injury films, the post external fixator application
410
film is the same as the injury film. That assumes the external fixation did not correct any deformity
411
to normal or the external fixator did not create any deformity other than what was present at time of
412
injury).
413
414
Reduction of a pelvic deformity necessitates an understanding of the three-dimensional deformity.
415
Diastasis in the anterior or posterior aspect of the pelvis in a mechanically unstable pelvis requires a
416
force vector medially along the x-axis. Simple internal rotation of the hemipelvis will close the
417
diastasis at the symphysis, but likely worsen the diastasis posteriorly. The surgeon prevents this
418
diastasis posteriorly and the flexion/internal rotation deformity by maintaining neutral rotation of the
419
hemipelvis while directing the reduction force posterior medially. Having good fixation in each
420
ilium, the surgeon compresses the two ilia while maintaining rotation of the hemipelvis.24
Skeletal Deformity Pelvis X-Fix
21
421
Furthermore, in line traction or traction at 45º often helps reduce the unilateral cephalad displaced
422
hemipelvis.
423
424
The measurement techniques (described in the Material and Methods section) help the surgeon
425
define the three-dimensional deformity of the hemipelvis. This allows for better reduction with the
426
external fixator and subsequent preoperative planning, thus making the definitive open reduction
427
internal fixation easier. These measurement techniques have been applied to the operative treatment
428
of pelvic malunion and nonunion patients.17. Measuring the rotation of the constant quadrilateral
429
plate to the plane of the sacrum (normal side internal rotation averages 6, ranges 0 to 19) and the
430
distance of the quadrilateral plate from the center on the CT scan allows the surgeon to separate the
431
internal rotation deformity from the adduction deformity. Flexion/extension deformity can be
432
determined by the following: (1) from the lateral three-dimensional CT scan, or (2) by comparing the
433
relationship between the ischial spine to the obturator foramen on the outlet view (the more caudad
434
the spine is in the obturator foramen the more flexion of the hemipelvis), or (3) from the obturator
435
acetabular line crossing of the teardrop (the more cephalad the crossing the more flexion).
436
437
Given the three-dimensional deformity of the hemipelvis, all the measurements are interrelated
438
which creates certain difficulties in defining the deformity. Once learned, these anatomical
439
interrelations allow a greater understanding of the actual deformity (i.e., adduction of the hemipelvis
440
increases the apparent flexion on the AP view). However, there are other problems that arise when
441
defining the deformity. In this series a potential difficulty included those patients with associated
442
acetabular fractures. The presence of fractures can prevent true measurements of deformities distal
Skeletal Deformity Pelvis X-Fix
22
443
to the fracture, i.e., rotation of the hemipelvis on the CT scan or the position of ipsilateral rami in
444
symphyseal disruptions. Another problem was false deformities of the hemipelvis. This can occur
445
with impactions, i.e., an impacted sacrum can look like an abduction deformity where in reality the
446
sacrum is impacted and the iliac wing springs back to its original position leaving a widened SI joint
447
on that side.
448
449
The actual (true?) deformity of the hemipelvis was similar pre-external and post-external fixation.
450
(Were you able to say after the definitive surgery of the acute injury or after the treatment of the
451
malunion that the preceding statement is highly accurate? In other words did those surgeries
452
reinforce this statement, “The actual deformity of the hemipelvis was similar pre-external and post-
453
external fixation” and that no additional new, different, or previously unrecognized deformities were
454
found at time of surgery? It is the word “similar” that stimulates the above questions. Does “similar”
455
mean “most of the time” or “in the majority of cases”, or what in terms of deformities or numbers of
456
patients?). Therefore, (it is our opinion?) the accident, not the external fixator caused the deformity.
457
However, as previously stated, in most cases the deformity worsened with the placement of the
458
external fixator. The most common deformities included cephalad and posterior translations,
459
internal rotation and flexion. These deformities are similar to the deformities found in our study of
460
the surgical treatment of pelvic malunions and nonunions.17
461
common deformities were internal rotation (67%) and cephalad (100%) and posterior (67%)
462
translation. The failure of external fixation and traction in definitive management of unstable pelvic
463
fractures explains the similarity in deformities between the acute deformities measured in this paper
464
and the deformities seen in our study of pelvic malunions and nonunions.17
In this malunion study the most
Skeletal Deformity Pelvis X-Fix
465
466
467
23
Skeletal Deformity Pelvis X-Fix
24
468
CONCLUSION
469
In this group of hemodynamically unstable patients with a mechanically unstable pelvis, we found a
470
73% worsening of rotational deformities (internal rotation and/or flexion) and the 67% worsening of
471
translational deformities (cephalad translation and/or diastasis) encountered posteriorly with the
472
use of pelvic external fixators. The use of the anterior pelvic external fixator in the open book
473
pelvis (with some posterior ligament stability) in a hemodynamically unstable patient appears to be
474
technically less complicated. When complete posterior disruption occurs, the application of an
475
anterior external frame may worsen the pelvic deformity and increase posterior displacement. The
476
biomechanical strength of an anterior external fixator for posteriorly disrupted pelvic injuries have
477
been questioned.14; however its use as a resuscitative frame is still supported. 4,8,10.12,13,15,19,23,24,28
478
Whether the posterior stabilizing external fixator will benefit patients with complete dissociation of
479
the hemipelvis is still undergoing investigation.5,6,9,29 Although speculative, the inability of some
480
external fixators to control hypotension in trauma patients may be in part, the malreduction
481
that remains after external fixation placement. In conclusion, the surgeon placing an external
482
fixator for a completely unstable hemi-pelvis must have a thorough knowledge of the 3-D
483
deformity of the pelvis and the force vectors required for reduction to prevent potential
484
worsening of the deformity and hemo-dynamic status. (You have the numbers for making the
485
statement regarding potential worsening of the deformity with anterior external fixation. However
486
your statement regarding malreduction possibly contributing to hemodynamic instability rather than
487
helping it, remains speculative and cannot be used as a conclusion unless there is personal
488
experience or a reference for it).
489
Skeletal Deformity Pelvis X-Fix
490
REFERENCES
491
1.
25
Beebe, KS, Reilly MC, Renard RL, Sabitino, et al. The effect of sacral fracture malreduction
492
on the initial strength of iliosacral screw fixation for transforaminal sacral fractures.
493
Presented at OTA 19th Annual Meeting in Salt Lake City Utah, October 2003.
494
2.
1996; 27 (2): B3-19.
495
496
3.
4.
Cryer H, Miller F, Evers B, Rouben L, Seligson D. Pelvic fracture classification: correlation
with hemorrhage. J. Trauma 1988; 28(7):973-79.
499
500
Bucholz R. The pathological anatomy of malgaigne fracture-dislocation of the pelvis. J.
Bone and Joint Surg., 1981; 63A(3):400-04.
497
498
Bircher MD. Indications and techniques of external fixation of the injured pelvis. Injury,
5.
Dickson KF, Coughlin R, Paiement G, Martin R, Klassen M. The Ganz Antishock. Pelvic
501
C-Clamp, Surgery of the Pelvis and Acetabulum: The Second International Consensus,
502
Pittsburgh, PA, October 21-27, 1994.
503
6.
Dickson KF, Matta J. Malreduction of pelvic fracture following external fixation, 30th
504
Annual Spring Meeting, Los Angeles Chapter, Western Orthopaedic Association, Santa
505
Barbara, CA, April 28-30, 1995.
506
7.
injuries using primary external fixation. Orthopaedic Transaction 1985; 9(3):434.
507
508
8.
511
Flint L, Babikian G, Anders M, Rodriguez J, Steinberg S. Definitive control of mortality
from severe pelvic fracture. Ann. Surg. 1990; 211(6):703-07.
509
510
Edwards C, Meier P, Browner B, Freedman M, Ackley S. Results treating 50 unstable pelvic
9.
Ganz R, Krushell R, Jakob R, Kuffer J. The antishock pelvic clamp. Clin. Orthop. 1991;
267:71-78.
Skeletal Deformity Pelvis X-Fix
512
10.
11.
12.
Gunterberg B, Goldie I, Slatis P. Fixation of pelvic fractures and dislocations. Acta Orthop.
Scand. 1978; 49:278-86.
517
518
Gruen G, Leit M, Gruen R, Peitzman A. The acute management of hemodynamically
unstable multiple trauma patients with pelvic ring fractures. J. Trauma 1994; 36(5):706-13.
515
516
Gilliland M, Ward R, Barton R, Miller P, Duke J. Factors affecting mortality in pelvic
fractures. J. Trauma 1982; 22(8):691-93.
513
514
26
13.
Huittinen V, Slatis P. Fractures of the pelvis - Trauma mechanism, types of injury and
principles of treatment. Acta Chir. Scand. 1972; 138:563-69.
519
520
14.
Kellam J. The role of external fixation in pelvic disruptions. Clin. Orthop. 1989; 241:66-82.
521
15.
Kellam J, McMurtry R, Paley D, Tile M. The unstable pelvic fracture. Operative treatment.
Orthopaedic Clinics of North America 1987; 18(1):25-41.
522
523
16.
Heidelberg, Springer-Verlag. 1993; 39.
524
525
17.
18.
19.
Moreno C, Moore E, Rosenberger A, Cleveland H. Hemorrhage associated with major
pelvic fracture: A multispecialty challenge. J. Trauma 1986; 26(11):987-94.
530
531
Matta JM, Saucedo T. Internal fixation of pelvic ring fractures. Clin Orthop 1989; 242:8397.
528
529
Matta JM, Dickson KF, Markovich GD. Surgical treatment of pelvic nonunions and
malunions. CORR 1996; 329:199-206.
526
527
Letournel E, Judet R. Fractures of the Acetabulum. Elson R, editor. Second Edition. Berlin
20.
Poole G, Ward E, Muakkassa F, Hsu H, Griswold J, Rhodes R. Pelvic fracture from major
532
blunt trauma. Outcome is determined by associated injuries. Ann. Surg. 1991; 213(6):532-
533
39.
Skeletal Deformity Pelvis X-Fix
27
534
535
21.
Rommens PM, Hessmann MH. Staged reconstruction of pelvic ring disruption: differences
536
in morbidity, mortality, radiologic results, and functional outcomes between B1, B2/B3, and
537
c-type lesions. J Orthop Trauma, 2002, 16:92-8.
538
22.
the pelvis. J Trauma 1983; 23:535-37.
539
540
23.
24.
Slatis P, Karaharju E. External fixation of unstable pelvic fractures: Experiences in 22
patients treated with a trapezoid compression frame. Clin. Orthop. 1980; 151:73-80.
543
544
Slatis P, Huittinen V. Double vertical fractures of the pelvis - A report on 163 patients. Acta
Chir. Scand. 1972; 138:799-907.
541
542
Semba R, Yasukawa K, Gustilo R. Critical analysis of results of 53 Malgaigne fractures of
25.
Sponseller P, Bisson L, Gearhart J, Jeffs R, Magid D, Fishman E. The anatomy of the pelvis
in the exstrophy complex. J. Bone and Joint Surg. 1995; 77-A(2):177-89.
545
546
26.
Tile M. Pelvic fractures – should they be fixed? J. Bone and Joint Surg. 1988; 70B:1-12.
547
27.
Tucker MC, Nork SE, Simonian PT, Routt ML. Simple anterior pelvic external fixation. J
Trauma. 2000 49:989-94.
548
549
28.
J. Bone and Joint Surg. 1982; 64-A(7):1010-20.
550
551
Wild J, Jr, Hanson G, Tullos H. Unstable fractures of the pelvis treated by external fixation.
29.
Witschger P, Heini P, Ganz R.
Beckenzwinge zur schockbekampfung bei hinteren
552
bechenringverletzungen. Applikation, biomechanische aspekte und erste klinische resultate.
553
Orthopade 1992; 21:393-99.
554
Skeletal Deformity Pelvis X-Fix
555
FIGURES
556
Figure 1
28
The three-axis system for defining the hemipelvis deformity. The x-axis defines the
557
flexion/extension rotational deformity and the diastasis/impaction translational
558
deformity. The y-axis defines the internal/external rotational deformity and the
559
cephalad/caudad translational deformity. The z-axis defines the abduction/adduction
560
rotational deformity and the anterior/posterior translational deformity.
561
Figure 2: 2A-2F
562
Twenty-six year old pedestrian vs. motor vehicle. Pedestrian sustained a left
563
sacroiliac dislocation and symphysis diastasis.
564
2A)
Anteroposterior view (AP) before application of external fixator.
565
2B)
Anteroposterior view (AP) after application of external fixator. Notice the
566
internal rotation of the left hemipelvis (widening of the ischium) and the
567
widening of the sacroiliac joint posteriorly while closing the symphysis
568
anteriorly.
569
2C)
After external fixator placement - computed tomography (CT) showing 52O
of internal rotation of the left hemipelvis.
570
571
2D)
Right hemipelvis with 1o internal rotation.
572
2E)
After open reduction internal fixation - AP.
573
2F)
CT scan showing 7o internal rotation of left hemipelvis and 1o internal
574
575
576
rotation of right hemipelvis.
Skeletal Deformity Pelvis X-Fix
577
Figure 3
29
The artist’s rendition of figure 2C with internal rotation deformity at 52º.
578
Furthermore, X being less than Y could suggest an adduction deformity or a medial
579
impaction.
580
581
Figure 4
The artist’s rendition of figure 2A with a parallel line to the sacrum and a 90º lines
582
measuring leg length discrepancy (X) versus sitting imbalance (Y). The width of the
583
ischium (2) increases with internal rotation.
584
585
586
Figure 5
5)
Pure internal rotation and adduction injuries with an inlet view.
5A)
Pure internal rotation injury. Notice the change in profile of the iliac wing
and obturator foramen.
587
588
589
590
591
592
593
594
595
596
597
598
5B)
Pure adduction injury. Notice the profile of wing and obturator foramen
remain constant.
Skeletal Deformity Pelvis X-Fix
599
600
601
602
603
604
605
30
Skeletal Deformity Pelvis X-Fix
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
Table 2:
31
DISPLACEMENT MEASUREMENTS DATA SHEET
Maximum Distance
Symphysis/Rami
Maximum
SI Diastasis
Maximum
Translation
Maximum Elevation of
symphysis (if > posterior
cephalad translation +FLEX/-EXT)
before __________
_________
_________
__________
after
__________
_________
_________
__________
CAUDAD
(Inlet)before
__________
_________
after
__________
_________
CEPHALAD
(Outlet)before __________
_________
_________
__________
AP
Maximum
translation
(+POS/-ANT)
Ischium Relation ilioischial
rotation to teardrop
(+IR/-ER)
(+IR/-ER)
Relation obturator
acetabular line to teardrop
(+FLEX/-EXT)
AP
before
_______
_________
_________
after
_______
_________
_________
CAUDAD
before _________
after
_________
ANTERIOR POSTERIOR
Discrepancy of ischial heights: ______ mm short side (circle one) right / left
Discrepancy of acetabulum dome: ______mm short side (circle one) right / left
Quadrilateral Plate
Distance to midline
(+ADD/-ABD)
CT Scan
before _________
Quadrilateral Plate
Rotation
(+IR/-ER)
PSIS/ASIS
Rotation
(+IR-ABD / -ER-ADD)
_________
_________
after _________
_________
_________
LEGEND: IR - internal rotation ER-external rotation
SI-sacroiliac joint
Flex - flexion
Ext-extension
Before- before external fixator
After-after external fixator
PSIS-posterior superior iliac spine
ASIS-anteriro superior iliac spine
Skeletal Deformity Pelvis X-Fix
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
LEGEND FOR TABLE 1
FINAL REDUCTION AFTER ORIF (Shouldn’t these be turned around re cms? The
excellent reduction is < 1 cm etc.
 E=EXCELLENT ( < 4 CM )
 G=GOOD ( 4 CM - 1 CM )
 P=POOR ( > 1 CM )
 3 STAGE = STAGE?D PELVIS RECONSTUCTION (Why use the number “3”
and tell us what “staged pelvis reconstruction means” and when it was done and
why it is different as opposed to the other patients who did not have this procedure?
FLEXION
 AC FX = ACETABULAR FRACTURE THAT PREVENTED DEFORMITY
MEASUREMENT
DEFORMITY AND ASSOCIATED INJURIES
 ABD = ABDUCTION
 ADD = ADDUCTION
 FLEX = FLEXION
 EXT = EXTERNAL EXTENSION
 IR = INTERNAL ROTATION
 ER = EXTERNAL ROTATION
 POST = POSTERIOR TRANSLATION
 CEPH = CEPHALAD TRANSLATION
 FX = FRACTURE
 R = RIGHT
 L = LEFT