ABSTRACT
Eighteen patients with unstable pelvic fractures,
treated from 1998 to 2003 were retrospectively reviewed. The
mean patients’ age was 39 years (range 17 to 79). Thirteen
patients were men and 5 women. The commonest cause was a road
traffic accident (N=14). There were 6 type-C and 12 type-B
fractures according to Tile’s classification. One of the
fractures was open. Fifteen patients sustained additional
injuries; four patients had lumbosacral plexus involvement.
External fixation as a definite management was applied to one
fracture; surgical reduction was scored from the post-operative
radiographs and Majeed’s score was used to assess the clinical
outcome. The mean follow-up was 51.7 months (range 15 to 77
months). Nine patients required admission to ITU
post-operatively. One patient developed superficial infection.
The mean hospital stay was 27 days (range 5 to 50 days). All the
patients had either good or excellent radiological reduction,
however 5 of them had a fair functional outcome according to
Majeed’s criteria. One patient died 2 years after surgery from
causes unrelated to pelvic injury and we were unable to trace 2
patients. One patient with type-C3 fracture underwent hip
resurfacing, 2 years following surgery. Four patients
complained of pelvic pain on mild exertion and two had sexual
dysfunction. Seven patients returned to their original job,
however the level of pain affected their performance
accordingly.
The
management of unstable pelvic fractures is challenging; initial
haemodynamic stabilisation takes preference and significantly
reduces mortality. Associated injuries are common and they might
further compromise the functional outcome. Open reduction and
internal fixation yields a satisfactory outcome, however
rehabilitation period is prolonged.
Keywords:
Pelvic fractures, unstable fractures, clinical outcome.
J.Orthopaedics 2005;2(2)e2
Introduction
Pelvic fractures are relatively rare injuries;
their incidence in trauma patients is quoted to range between 3
% and 8.2 % and instability occurs in 13 % to 17 % of cases (1).
Because of the large force that is required to
disrupt the pelvis, pelvic fractures are indicative of
high-energy transfer to the patient and therefore, often
combined with other injuries (1). Road Traffic Accidents (RTA)
is the commonest cause of major pelvic ring injuries, followed
by falls from a height, athletic or crush injuries (2). The
mortality rate of haemodynamically unstable patients with pelvic
fractures may be as high as 40 to 50 %, therefore prompt
resuscitation that may involve several specialties is essential
(2). Open reduction and internal fixation of the unstable pelvic
ring fractures has been suggested to provide the best stability
of fixation and the best clinical outcome (7,8), however
rehabilitation period is prolonged.
The purpose of this study was to evaluate the
morbidity and functional outcome of surgically treated unstable
pelvic fractures in our department.
Patients and
method
A retrospective study was conducted on eighteen
patients who underwent surgical intervention for unstable
fractures of the pelvic ring between 1998 and 2003. Instability
was defined according to Tile’s classification (3).
The case notes were studied and demographic and
clinical data was recorded, which included; age, gender,
mechanism of injury, type of fracture, associated injuries,
definitive surgery, postoperative ITU admission, hospital stay
and complications.
The quality of surgical reduction was scored from
the post-operative radiographs by implementing the scoring
system that was introduced by Pohlemann et al (9).
The clinical outcome at the time of follow up was
assessed via the scoring system introduced by Majeed (10); the
current job status was also recorded.
Nine patients were transferred directly to our
Accident and Emergency Department and the remaining nine
patients were referred from surrounding trusts, as our unit acts
as the referral centre for these injuries. Four of the nine
patients who were transferred directly to our hospital arrived
with signs of hypovolaemia; following initial fluid
resuscitation and stabilization of the pelvic ring with an
external fixator at the A&E department, three of them underwent
an urgent laparotomy following urgent abdominal ultrasound which
indicated free fluid in the peritoneal cavity. Surgical
stabilisation of the pelvic ring with a combination of
techniques was carried out as an emergency procedure within 6
hours of arrival in seven patients; the remaining two patients
underwent surgery within 24 hours of admission. The remaining
nine patients were referred by surrounding hospitals within an
average of 4 days from the time of their injury and presented
with stable vital signs on admission to our Unit. The ISS of
patients who were directly transferred to our A&E had a mean
value of 30 (range 16-62).
The mean patients’ age was 39 years (range 17 to
79 years). Thirteen patients were men and five women. Road
traffic accidents were the cause of pelvic fracture in 14
patients, followed by fall from a height in 3 patients and
injury from a falling object in one patient. There were 6 type-C
and 12 type-B fractures according to Tile’s classification (3).
One of the fractures was open communicating with the vagina.
Two patients who underwent urgent laparotomy were found to have
a ruptured spleen. Four patients had urological injuries; two
had sustained bladder tears and the other two kidney lacerations
resulting in one nephrectomy. Nearly half the patients had
associated long bone fractures. One patient with comminuted
fracture of the humerus had additional rupture of the axillary
artery and brachial plexus injury; the damage to axillary artery
was successfully repaired but the brachial plexus had only
partially recovered at the time of follow up. Neurological
injury of the lumbosacral plexus was identified in four patients
at the initial evaluation; 3 were combined motor and sensory
deficits affecting the L4, L5, and S1 nerve roots and the
remainder was an isolated sensory deficit of the S1 and S2 nerve
roots. (Table 1)
Nine patients had fixation of the symphysis pubis
as well as the sacro-iliac (SI) joint. Three patientsunderwent
internal fixation of the acetabulum and sacro-iliac joint and
another three had fixation of the sacro-iliac joint that was
combined with external fixation in two patients. One patient had
fixation of the symphysis pubis whereas in another one that was
combined with external fixation. External fixation as definitive
treatment was applied for eight weeks in one patient. All
patients received Cefuroxime perioperatively. Prophylaxis
against Deep Venous Thrombosis included low molecular weight
heparin (Enoxaparin 20mg sc) for the time of hospitalisation and
thrombo-embolic deterrent stockings for six weeks following
surgery. Seven patients required admission to ITU
postoperatively with an average stay of 3.5 days (range 2-7
days). The average stay in hospital was 27 days (range 5 to 50
days). (Table 2)
The mean
follow-up was 51.7 months (range 13 to 77 months). The data
regarding the clinical outcome were collected via a
questionnaire that was posted to patients along with a pre-paid
self-addressed envelope. Those who did not reply were conducted
by phone. Analysis of the data was performed in Excel 2002.
Table I
No (N=18) |
Age |
Gender |
Mechanism of injury |
Type of Fracture |
Neurological
injury |
Associated injuries |
1 |
27 |
F |
RTA |
B2 |
S1, S2,
sensory |
Retroperitoneal
bleed,
Vaginal laceration,
# radius, # ankle |
2 |
29 |
M |
RTA |
C3 |
|
# mandible, # os
calcis |
3 |
44 |
M |
Fall From hgt |
B2 |
|
Head injury, # tibia |
4 |
38 |
F |
Falling object |
B2 |
|
# 2
ribs, splenic laceration |
5 |
40 |
F |
RTA |
B2 |
L5, S1 |
Bladder tear, #
forearm |
6 |
17 |
M |
RTA |
B2 |
|
|
7 |
55 |
M |
RTA |
C3 |
|
Kidney contusion, #
forearm (open) |
8 |
24 |
M |
RTA |
B3 |
|
Intracerebral
bleed, flail chest |
9 |
52 |
M |
RTA |
C1 |
L4, L5 |
Kidney laceration,
splenic laceration, # wrist, |
10 |
43 |
M |
Fall From hgt |
B1 |
|
|
11 |
40 |
M |
RTA |
B3 |
|
# humerus, #
tibia, axillary artery injury,
brachial plexus injury. |
12 |
39 |
M |
Fall From height |
B2 |
|
# ribs, Pneumothorax,
# humerus, # femur |
13 |
35 |
M |
RTA |
C3 |
|
# femur, # ankle |
14 |
41 |
F |
RTA |
B3 |
|
Ruptured urethra |
15 |
42 |
M |
RTA |
C3 |
L5, S1 |
Bladder tear |
16 |
36 |
M |
RTA |
B3 |
|
# 2 ribs,
Pneumothorax |
17 |
79 |
F |
RTA |
B2 |
|
# humerus, # wrist |
18 |
21 |
M |
RTA |
C1 |
|
# calcaneous, head
injury |
TABLE II
Serial No (N=18) |
Type of fracture |
Type of fixation |
ITU stay
(days) |
Hosp. Stay
(days) |
Complications |
Radiological score |
Clinical outcome |
1 |
B2 |
External fixation,
ORIF symphysis pubis |
7 |
31 |
Nil |
Excellent |
Good |
2 |
C3 |
Fixation SI joint,
Acetabular Recon. |
2 |
39 |
Nil |
Good |
Excellent |
3 |
B2 |
Fixation SI joint,
ORIF Pubic Rami |
|
7 |
Pneumonia |
Excellent |
Good |
4 |
B2 |
External Fixation |
|
14 |
Paralytic ileus |
Good |
Good |
5 |
B2 |
External fixation
Fixation SI joint |
|
28 |
Errectile
dysfunction |
Excellent |
Good |
6 |
B2 |
Fixation SI joint,
ORIF Pubic Symphysis |
2 |
18 |
Nil |
Good |
Excellent |
7 |
C3 |
Fixation SI joint
Acetabular Recon. |
|
20 |
Superficial
infection |
Excellent |
Fair |
8 |
B3 |
ORIF Pubic Symphysis |
4 |
25 |
Nil |
Excellent |
Excellent |
9 |
C1 |
Fixation SI joint,
ORIF Pubic Symphysis |
4 |
30 |
Paralytic ileus |
Good |
Fair |
10 |
B1 |
Fixation SI joint,
ORIF Pubic Symphysis |
|
5 |
Nil |
Excellent |
Excellent |
11 |
B3 |
Fixation SI joint,
ORIF Pubic Symphysis |
|
37 |
Small PE |
Good |
Good |
12 |
B2 |
External Fixation
Fixation SI joint |
4 |
39 |
MRSA at pin site |
Excellent |
Excellent |
13 |
C3 |
Fixation SI joint,
Acetabular Recon. |
|
50 |
Nil |
Good |
Fair |
14 |
B3 |
Fixation SI joint,
ORIF Pubic Symphysis |
4 |
40 |
Errectile
dysfunction |
Excellent |
Excellent |
15 |
C3 |
Fixation SI joint,
ORIF Pubic Symphysis |
|
38 |
Paralytic ileus |
Good |
Fair |
16 |
B3 |
Fixation SI joint,
ORIF Pubic Symphysis |
2 |
14 |
Nil |
Excellent |
Lost to f-up |
17 |
B2 |
Fixation SI joint |
3 |
32 |
Pneumonia |
Good |
Died 2y post-op |
18 |
C1 |
Fixation SI joint,
ORIF Pubic Symphysis |
|
35 |
Nil |
Excellent |
Lost to f-up |
Results
One
patient died 2 years following surgery from causes not related
to pelvic injury. We were unable to trace 2 patients; therefore
15 patients were available for follow up.
Complications: The commonest
postoperative complication was paralytic ileus, which occurred
in 3 patients, followed by pneumonia in 2 patients. One patient
developed small pulmonary embolus that was treated successfully;
there was not any clinically detected Deep Vein Thrombosis.
Bladder rupture occurred in two patients and was repaired
without sequelae, following transurethral catheterisation. Two
patients reported erectile dysfunction in the early
post-operative period; one of them had associated urethral
damage. Both patients regained normal erectile function within a
year following their injury. All fractures united and there were
not implant failures. Two patients developed superficial
infection, which were successfully treated by oral antibiotics.
There were no deep infections.
Radiological results: Two patients
with type-C fractures had excellent radiological reduction with
anterior displacement of symphysis pubis of less than 5mm, while
the remaining 4 patients had good radiological reduction with
residual displacement of symphysis pubis between 6mm and 10mm.
Eight patients with type B fractures had excellent radiological
score and the remaining four had good radiological reduction.
Clinical outcome: According to
Majeed’s score, excellent and good clinical outcome was achieved
in 6 and 5 patients respectively. Four patients had a fair
result with activity related intense pain being the main
complaint; all had initially sustained a C-type injury and
eventually had to change their original jobs. Four patients
complained of a slight limp and pain on mild exertion. Two
female patients complained of dyspareunia; they both had
residual displacement of the symphysis pubis, of more than 5mm,
following surgical reduction and one had also sustained a
vaginal laceration at the time of injury. One patient with an
associated acetabular fracture (C3-type) underwent hip
resurfacing arthroplasty 2 years after surgery, due to intense,
continuous pain at rest with radiological evidence of
osteoarthritis.
Outcome evaluation: The study
revealed that radiological result was superior to the clinical
result and that residual anterior displacement of symphysis
pubis beyond 5mm following fixation is likely to lead to
residual symptoms. However, this could not be statistically
proved due to small number of patients that were reviewed. All
neurological injuries involving the lumbosacral plexus were
fully recovered at the time of follow-up. The brachial plexus
injury however, recovered only partially. Seven patients with
B-type fractures were pain-free at the time of study as opposed
to one patient with C-type fracture. Seven patients returned to
their original job, however the level of pain influenced their
performance accordingly.
Discussion
Pelvic fractures are relatively rare injuries;
their incidence in trauma patients is quoted to range between 3
% and 8.2 % and instability occurs in 13 % to 17 % of cases (1).
The commonest cause of a pelvic ring disruption is a Road
Traffic Accident (2) and that involved 70% of our patients.
Because of the large force that is required to disrupt the
pelvis, pelvic fractures are indicative of high-energy transfer
to the patient and therefore, often combined with other injuries
(1,17).
Patients with major pelvic injuries need a
multidisciplinary approach and should be treated in
well-equipped and staffed centres. Stabilisation of vital
parameters is the first goal, and a standardised trauma protocol
for diagnostic policy as well as for surgical treatment should
be followed routinely (17). Associated intra-abdominal injuries
should be evaluated by emergency abdominal ultrasound to
diagnose or exclude intra-abdominal bleeding (17). Ultrasound
was sufficient to diagnose intra-abdominal haemorrhage in both
our patients with splenic lacerations, however laparotomy was
unnecessary in the third patient.
Haemodynamically unstable patients with pelvic
fractures have a mortality of 40 to 50% and this increases
markedly if there is associated head injury (2). Some authors
attribute this high mortality to exsanguinating haemorrhage
within the pelvis, others are for the opinion that massive
bleeding from pelvic fractures is uncommon and that mortality is
related to associated injuries (4). External fixation can be
applied in 15 minutes by an orthopaedic surgeon or a physician
credentialed in this procedure (18). This method reduces the
relative volume of a fractured pelvis, thereby reducing the
potential space for haemorrhage. If the patient remains unstable
despite the resuscitation efforts, should undergo angiographic
embolisation or be brought to operating theatre for surgical
intervention (18). None of the 4 haemodynamically unstable
patients who were transferred to our A&E department required
angiographic embolisation, however three of them underwent an
urgent laparotomy following initial stabilization of the pelvic
ring by an external fixator. Two were found to have sustained
splenic lacerations and in the third a retroperitoneal haematoma
was burst into the peritoneal cavity. The vital signs of the
fourth patient were stabilised after application of a pelvic
sling in the A & E department.
The incidence of an open, unstable pelvic ring
disruption with connection occurring between the fracture and
skin, rectum or vagina is 3.5 to 4.5% (1). Brenneman et al (ref)
reported that morbidity and mortality of the patients with open
pelvic fractures were higher than in patients with closed
injuries and Kobak et al (ref) reported death due to sepsis in a
patient with open fracture. One female patient of ours had an
open fracture that was communicating with the vagina; that was
treated successfully without complications.
The mortality rate in this small series of
unstable pelvic fractures was zero; the fact that the only
patient with serious brain injury recovered fully and the open
fracture was communicating with the vagina rather than the
rectum, as well as the prompt multidisciplinary approach of
haemodynamically unstable patients may explain this favourable
outcome, which comes in contradiction with the recent literature
( 17).
Associated injuries occurred in most of our
patients (15 out of 18 patients). Corovessis et al reported
that concomitant injuries have negatively affected the clinical
outcome of unstable pelvic fractures despite the good
radiographic score of surgical reduction. The presence of
associated injuries has certainly increased morbidity in our
group of patients and negatively affected the functional outcome
in the case of a brachial plexus injury, a calcaneal fracture
and a vaginal laceration. Four out of eighteen patients (22%)
had an associated acetabular fracture and one patient underwent
a hip resurfacing for progressed hip osteoarthritis within 2
years of injury.
The incidence of neurological injuries in this
study was 22% which is comparable with that reported in the
literature. Three patients with combined motor and sensory
neurological deficits affecting the L4, L5, S1 nerve roots were
fully recovered at the time of follow-up. One patient with
isolated sensory deficit of S1 and S2 nerve roots was still
complaining for perineal dysaesthesia and dyspareunia, 29 months
following injury. The small number of patients in this study may
be a reason for the most favourable potential for recovery of
the neurological injury at the L4, L5, and S1 nerve roots, as
compared with the literature (10,20). Early rigid stabilisation
of both anterior and posterior pelvic ring injury, which is what
was performed in our patients, has been suggested as a potential
reason for favourable prognosis of these injuries.
Stabilisation of the unstable pelvic ring
injuries can be achieved by external and/or internal fixation
(17). External fixation can be applied fast resulting in reduced
haemorrhage in the intrapelvic space by tamponade (18). Open
reduction and internal fixation of the unstable pelvic ring
fractures provides the best stability of fixation as well as
best clinical outcome (7,8). Unstable type-B1 injuries should be
fixed (17) whereas type-B2 and B3 fractures could be treated
non-operatively, because the pelvis has the elasticity to
restore to a near normal position (19).
Some of the reported long-term morbidity of an
unstable pelvic ring fracture include chronic pelvic pain,
impaired function of the pelvis for sitting and weight bearing
that results from pelvic non-union, pelvic malunion and leg
length discrepancy (6).
Non-union or leg length discrepancy of more than
1cm did not occur in our patients but displacement of symphysis
pubis of more than 5mm was associated with residual symptoms
such as pain and dyspareunia in 2 females. The small number of
patients did not allow us to derive statistically significant
correlation between the two parameters.
Our study shows that the antibiotic prophylaxis
used in the unit is effective as only one patient developed
superficial infection. The deep venous thrombosis (DVT)
prophylaxis is also effective as no patient developed clinical
deep venous thrombosis.
Even though the review of literature shows
several studies have supported this both in biomechanical
studies and clinical trials there is a lot of controversy about
the long-term outcome of unstable pelvic injuries. Berner in
1982 reported a rate of 16% unsatisfactory functional, and 17%
unsatisfactory radiological result in a group of 42 patients
treated non-operatively after combined disruptions of the pubic
symphyses and the sacro-iliac joint. (11) After treatment of a
similar injury with open reduction and internal fixation, the
rate of unsatisfactory functional rating was zero and 10% in the
radiological rating.
Rargnarsson in 1993 reported on 21 patients after
SI joint disruption treated with internal fixation using plating
(12). He stated the radiological position of the pelvic ring as
“unchanged to the post-operative radiographs”, a rate of 14%
poor functional results were reported.
One of the reasons for the poor result previous
studies have postulated is that a residual displacement of 10mm
and more was critical for a significant increase of residual
pain (13). Semba et al also reported a correlation of primary
anterior and posterior displacement exceeding 10mm being
correlated with a markedly higher incidence of severe low back
pain. (13) Holdsworth in 1948 reported that 50% of the patients
they studied returned to their original job (14). Our study
showed that 7 patients returned to their original jobs. In the
largest series of patients treated with open reduction and
internal fixation of unstable posterior pelvic injuries, 67%
returned to their former jobs without restrictions (15). Another
study where all fractures were reduced operatively to less than
10mm of residual displacement; 35% of patients had neurologic
injuries, and another 23% had associated injuries inhibiting
normal gait. Females with pelvic fractures tended to have
increased urinary complaints and dyspareunia, which were shown
to correlate with residual displacement of >5mm. (16)
Our study
has shown that anatomical restoration of the pelvic ring
correlates with higher probability of a good clinical outcome.
Conclusions
Pelvic
fractures are challenging injuries to manage. Stabilisation of
vital parameters takes preference and significantly reduces
mortality. Associated injuries are common and often have a
substantial effect on the patient’s psychological status.
Rehabilitation period is prolonged; however proper management
yields a satisfactory outcome. Further analysis and studies
including a larger number of patients are required to identify
the prognostic factors for the late sequelae. This preferably
should be a valid statistical analysis of outcomes of patients
treated nonsurgically, by external fixation or by internal
fixation in a randomised prospective study with standardised
indications for surgery. A standardised trauma protocol for
diagnostic policy as well as for initial and definitive surgical
treatment should be followed routinely.
References
1)
Mucha P, Farnell M. Analysis of pelvic
fractures management. J.Trauma, 1984; 24(5): 379-386
2) Moreno C,
Moore EE, et al: Haemorrhage associated with major pelvic
fracture: A multispeciality challenge. J Trauma 26:987-994,1986
3) Tile M.
Pelvic ring fractures: should they be fixed? J Bone Joint Surg
1988; 70B:1
4) Poole Gv,
Ward E F, Causes of mortality in patients with pelvic fractures.
Orthopaedics 1994; 17:691-6
5) Mears DC, Fu
F: External fixation in pelvic fractures. Orthop clinic north Am
1980; 11:465-79
6) Smith J,
Western journal of Medicine. San Francisco: Feb 1998; 168:124-6
7) Matta J,
Saucedo T. Internal fixation of pelvic ring fractures. Clin.
Orthop 1989:242:83-97.
8) Tile M. Fractures of
the pelvis and acetabulum. Baltimore: Williams and Wilkins, 1984
9) Pohlemann T
et al. Outcome after pelvic ring injuries. Injury 1996; 27
Suppl2: B31-8
10) Majeed SA.
Grading the outcome of pelvic fractures. JBJS Br.1989 Mar;
71(2): 304-6
11)Berner W,
Oestern H-J 1982; 85:377-387
12) Rargarsson
B, Oleurd C. Anterior square-plate fixation of sacroiliac
disruption. Acta Orthop.Scand.1993;64(2):138-142
13) Semba R,
Yasukawa K, Gustillo R Critical analysis of results 53 Malgaigne
fractures of pelvis.J.Trauma,1983;23(6):535-537
14) Holdsworth
FW: Dislocation and fracture dislocation of pelvis.J Bone Joint
Surg Br 30;461-466,1948
15)Tornetta P,
Dickson K, Matta JM: Outcome of rotationally unstable pelvic
ring injuries treated operatively.Clin Orthop 39:147-151,1986
16) Copeland
CE,Bosse MJ, McCarthy ML: Effect of trauma and pelvic
fracture on female genitourinary, sexual and reproductive
function.J Orthop Trauma 11:73-81,1997
17) Van Veen IH,
van Leeuwen AA, van Popta T, van Luyt, Bode PJ, Vugt AB:
Unstable pelvic fractures: a retrospective analysis. Injury.1995
Mar; 26(2): 81-5
18) Coppola T,
Coppola M: Emergency department evaluation and treatment of
pelvic fractures. Emerg Med Clin of North Am.2000 Feb; 18(1):
1-27,
19) Tile
M.Pelvic ring fractures: should they be fixed? J Bone Joint Surg
1988;70B: 1
20) Helfet DL,
Koval KJ, Hissa EA et al:Intraoperative somatosensory evoked
potential monitoring during acute pelvic fracture surgery. J
Orthop Trauma 9:28-34,1995
|