Session I - Pelvis

Vertically Unstable Pelvic Fractures Fixed with Percutaneous Iliosacral Screws: Does Posterior Injury Predict Fixation Failure?

Damian R. Griffin, MA, FRCS (Orth); Adam J. Starr, MD; Charles M. Reinert , MD; Alan L. Jones, MD; Shelly Whitlock, CCRA; University of Texas Southwestern Medical Center, Dallas, TX

Introduction: Percutaneous iliosacral screws have a developing role in the management of pelvic fractures, but little evidence has been reported to guide their use in vertically unstable fractures This study was performed to test the hypothesis that vertically unstable pelvic fractures (OTA type 61-C) in which the posterior injury is a vertical fracture of the sacrum (OTA type 61-C1.2) are more likely to fail after closed reduction and percutaneous iliosacral screw fixation than those with dislocations or fracture dislocations of the sacroiliac joint (OTA type 61-C1.3).

Materials and Methods:

Inclusion criteria

Patients admitted with an unequivocal vertically unstable pelvic fracture (defined as at least 1 cm of combined vertical displacement on initial inlet and outlet views, or SI dislocation with at least 1cm of separation of the entire SI joint on initial CT, or a complete vertical sacral fracture with at least 1cm of separation throughout its area on initial CT) and who were treated with closed reduction and percutaneous iliosacral screws for the posterior injury were included.

Exclusion criteria

Patients with any evidence of a lateral compression component to their injury, (defined as a sacral impaction fracture or internal rotation of the affected hemipelvis on CT), patients whose treatment was delayed by more than 3 weeks, patients with 'H'-shaped sacral fractures, patients whose initial posterior stabilization utilized a method other than iliosacral screws, or patients whose initial posterior surgery was performed at another hospital were excluded.

Data collection

All patients with pelvic fractures admitted between January 1, 1993 and December 31, 1998 were identified from the trauma registry. Hospital records and operative reports were examined to identify those treated with iliosacral screws. Of these patients, initial anterior posterior, inlet and outlet radiographs, and computed tomography scans were examined to identify those that were unequivocally vertically unstable. Demographic and injury information was collected from the trauma registry. Immediate postoperative and follow-up AP, inlet and outlet radiographs from a minimum of 12 months post-injury were examined. Position, length and numbers of iliosacral screws, and any evidence of metalwork failure (e.g., bending or breakage) were recorded. Residual postoperative displacement and late displacement of the posterior pelvis were independently measured on each view by 2 observers. Differences in measured displacement of 0.5cm or more were resolved by consensus.

Analysis

The main outcome measure was failure, defined as at least 1cm of combined vertical displacement of the posterior pelvis compared to immediate postoperative position. The main analysis was for association between fracture pattern and failure. Patient demographic data, iliosacral screw position, and anterior pelvic fixation methods were also studied. Fisher's exact test was used to identify association between variables, and P<0.05 was considered significant.

Results:

Patients

Eight hundred sixteen pelvic fractures were identified, of which 95 included widely separated or vertically displaced posterior injuries; 21 had evidence of lateral compression and were excluded. Two patients had double vertical or 'H'- shaped sacral fractures, and one had an unclassifiable complex sacral and pelvic fracture. Six patients died (5on the day of injury and one 36 hours after surgery). Surgery was delayed more than 3 weeks in one patient (posterior open reduction and plate fixation 5 weeks post injury) and one patient was referred from another hospital after failed anterior sacroiliac joint plating. The radiographic record was incomplete for one patient. Thus, the cohort was made up of 62 surviving patients with unequivocally vertically unstable pelvic fractures in whom the posterior injury was treated with closed reduction and percutaneous iliosacral screw fixation. Thirty-two of these were dislocations or fracture dislocations of the sacroiliac joint (OTA type 61-C1.2), and 30 were vertical fractures of the sacrum (OTA type 61-C1.3). Sixteen of 62 patients were women; the mean age was 34 (range17 ­ 59) and mean ISS was 20 (range 9 ­ 50).

Surgery

Pelvic surgery was performed within 5 days in 53/62 patients. Reduction was to within 1 cm or less of combined displacement on inlet and outlet views in all but 4 patients. Two patients were fixed with a single screw in S1, 3 with 2 screws in S2, 56 with a screw in each of S1 and S2, and one with 2 screws in S1 and one in S2. The longest screw reached the sacral body in 18 patients, the level of the far foraminae in 15, the far ala in 22 and across the far sacroiliac joint in 7. Anterior fixation was by symphysis plate in 27 patients, percutaneous superior ramus or anterior column screw(s) in 19, external fixation in 10, and was not performed in 6. Patients were managed with foot flat weight bearing for 3 months after surgery and were then allowed to advance to full weight bearing.

Outcome

Fixation failed in 4 patients, all with vertical sacral fractures. In the first patient, the S1 screw bent, the S2 screw loosened, and the symphysis plate pulled off by one week. In the second patient, the S1 and S2 screws, and the symphysis plate loosened within 2 weeks. In the third patient, the S1 and S2 screws backed out, and the anterior external fixator failed by 3weeks. In the fourth patient, the S1 and S2 screws loosened and the fracture displaced between 2 and 3 weeks, but the right and left retrograde superior ramus screws remained in an acceptable position. All 4patients required revision fixation. In 2 further cases with vertical sacral fractures there was evidence that the fracture had only barely been held by the fixation. In each patient, one screw had bent and the other loosened, but these fractures had healed and the follow-up radiographs did not meet the displacement criteria for failure.

Statistical analysis

A vertical sacral fracture pattern was significantly associated with failure (Fisher Exact test: P=0.022); the excess risk of failure compared with sacroiliac joint injury was 6% (95% confidence interval 3% - 30%). There was no significant association between failure and anterior fixation method, iliosacral screw arrangement or length, or any demographic or injury variable.

Discussion: This paper reports the results of iliosacral screw fixation of the posterior injury in a consecutive series of pelvic fractures which are the most difficult to treatthose which are vertically unstable. Combined with appropriate anterior fixation, this method was successful in maintaining reduction in 94% of cases. In 6% of cases, fixation failure occurred and necessitated further surgery. These failures were significantly associated with fracture pattern, occurring only in those with a vertical sacral fracture.