Session VIII - Pediatrics/Spine

Clinical Outcomes in Skeletally Immature Unstable Pelvis Fractures

Paul S. Shurnas, MD; Wade R. Smith, MD; Gianna Luzko, BS; Gaia Georgopoulos, MD; Steven J. Morgan MD, University of Colorado Health Sciences, Denver, CO

Introduction: There are few studies in the literature that evaluate the natural history of unstable pelvis fractures in skeletally immature patients and the effect of operative or nonoperative treatment. Recent studies suggest that nonoperative therapy may result in pelvic asymmetry leading to poor clinical results. The purpose of this study was to determine the natural history of unstable fractures and what factors influence clinical and functional outcomes.

Materials and Methods: We reviewed the clinical results, radiographs and computed tomography scans of all patients with open triradiate cartilage and an unstable pelvis (OTA- Type B and C) injury admitted from 1986 to 1998 at 2 level I trauma centers. Tile and OTA classification systems were used, and Modified Injury Severity Score (MISS) was assessed. Pelvic asymmetry was determined on the anteroposterior radiograph by the method of Keshishyan, et al. Patients meeting the inclusion criteria were re-evaluated by the authors with a standardized physical examination, radiographs (inlet/outlet, anteroposterior, Judet) and Short Musculoskeletal Function Assessment Questionnaire (SMFA). Patients with closed triradiate cartilage and acetabular fractures without pelvic ring disruption were excluded.

Results: Fifteen patients were identified with unstable fractures and open proximal physes. These 15 fractures represented 10% of all skeletally immature pelvis injuries over the study period as determined from the trauma registry. There were 4 type B pelvis fractures, and 11 type C pelvis fractures (6 with associated acetabular fractures) included in the study. Twelve of 15 patients (80%) were available for re-evaluation at the follow-up examination. Three patients deported to Mexico with an average of 1 year of follow-up were not available for re-evaluation. The mean age at the time of injury was 8.9 years (range 2.7-12.8). Follow-up averaged 7.1 years (range 1.4-14.6).

The 4 patients with Type B fractures had a mean MISS of 26.4 (range 11-50). Three of 4 patients (75%) required a stay in the intensive care unit (ICU) (mean stay 5.3 days). One of 4 patients (25%) required a blood transfusion (3 units) and embolization of an arterial pelvic bleed. Two patients had a MISS >25 with an average ICU stay of 9.5 days, and 2 patients had a MISS < 25 with an average ICU stay of 1 day. Three patients were treated nonoperatively (all Type B1), and 1 was treated by external fixation without posterior fixation (Type B3). The pelvic asymmetry for the nonoperative group averaged 1 cm and 1.3 cm for the patient treated by anterior external fixation at follow-up. The average time to weight bearing was 46 days (range 20-67) in the nonoperative group and 30 days in the patient treated by external fixation.

The 11 Type C fractures had a mean MISS of 30.5 (range 11-50). Ten of 11 (91%) patients required an ICU stay (mean stay 8 days). Six of 11 (55%) patients required blood transfusions (mean of 6.3 units) and 3 had embolizations. Five patients (45%) had a MISS < 25 and a mean ICU stay of 0.8 days, and 6 patients (55%) with a MISS > 25 had a mean ICU stay of 14.3 days. Nine of 11 patients were treated operatively by external fixation with or without posterior fixation. The 2 nonoperatively treated patients had an average of 1.3 cm of pelvic asymmetry at follow-up. Six patients were treated with anterior and posterior fixation, and their pelvic asymmetry averaged 0.8 cm at follow-up. The 3 patients treated solely with anterior external fixation had an average of 2.7 cm of pelvic asymmetry at follow-up. Operatively treated patients began weight bearing an average of 20 days sooner than nonoperative patients did. The mean time to weight bearing was 48.9 days (range 28-107).

At the follow-up examination, patients with less than 1cm of pelvic asymmetry had no lumbar or sacroiliac pain, no or mild sacroiliac tenderness, no Trendelenburg gait or lumbar scoliosis, and lower bother and dysfunction scores on the SMFA compared to patients with > 1 cm of pelvic asymmetry. All patients with more than 1cm of pelvic asymmetry had one or more of the following: non-structural scoliosis, lumbar pain, Trendelenburg gait, or complaints of sacroiliac joint tenderness and pain. Patients with a MISS < 25 had lower bother and dysfunction index scores on the SMFA compared to patients with a MISS > 25. Pelvic asymmetry evaluated on radiographs from the time of injury through the follow-up exam did not remodel in any patient.

Operative morbidity included 2 pin-site infections that were resolved with cleaning and oral antibiotics, 1 wound infection resolved with debridement and 3 days of intravenous antibiotics, and 1 symphysis osteomyelitis after open reduction and internal fixation (ORIF) that was resolved with debridement and 6 weeks of intravenous antibiotics. There were no fatalities.

Discussion: A MISS less than 25 was associated with better results in terms of strength, range of motion, reflexes, lower ICU stays and SMFA score. Pelvis fractures with greater instability regardless of MISS required more blood transfusions. Posterior fixation (sacroiliac screws or ORIF) combined with anterior fixation (external fixation or ORIF) resulted in less than 1 cm of pelvic asymmetry in all patients treated in this fashion. Three patients treated nonoperatively (2 B1 and 1 C1, with associated triradiate separation) had less than 1 cm of pelvic asymmetry at follow-up and good clinical as well as functional results.

Conclusion: Pediatric unstable pelvis fractures can have many associated injuries, hemodynamic instability that often requires blood transfusion, long-term morbidity and functional problems. Patients with less associated injuries (MISS < 25) and pelvic asymmetry < 1 cm had better clinical and functional results. Pelvic asymmetry did not remodel in any patient. Operative treatment appeared to decrease long-term morbidity and functional problems by maintaining pelvic symmetry but did not guarantee maintenance of pelvic symmetry in posterior ring injury unless posterior fixation was employed. As a result of this study we have instituted an unstable pediatric pelvis fracture protocol and a multi-center prospective study.