Symphyseal disruptions are classically the result of high-energy trauma, commonly requiring anterior plate fixation to achieve stability.1,2 The symphyseal cartilage is typically left in place, and compression across it is used to gain stability. The addition of posterior pelvic fixation is necessary to achieve stability in type 3 pelvic injuries characterized by a pattern of complete rotational and vertical instability.1 Previous studies have reported implant failure rates ranging from 12% to as high as 75% depending on the method of fixation and injury pattern. Implant failure is associated with recurrent symphyseal widening and loss of reduction with a reported range of 4%–24%.2–8 In 1 recent series, 93% of patients with implant failure had recurrent symphyseal widening of up to 23 mm (mean, 8.4 mm), which was statistically different than patients with no implant failure.3 Loosening of the construct without catastrophic failure may also allow for widening of the symphyseal space but may not have substantial implications on outcome.
Multiple factors have been correlated with failure of symphyseal reductions including the use of short plates and the quality of the reduction.8,9 However, we are unaware of any data examining the impact of symphyseal cartilage excision during open reduction and internal fixation (ORIF) on implant failure rates and postoperative symphyseal widening. Because friction is the primary biomechanical factor associated with stability when plating is used, we hypothesized that removal of the symphyseal cartilage would increase friction and create a more stable construct resulting in fewer implant failures and less postoperative displacement after symphyseal fixation. The purpose of this study was to compare the initial and final symphyseal alignment and incidence of implant failure with and without symphyseal cartilage excision after symphyseal reduction and fixation.
PATIENTS AND METHODS
We performed an IRB-approved retrospective review of the records of patients (2000–2013) treated at two level-one trauma centers with anterior posterior compression (APC) type 2 and 3 pelvic injuries as classified by the system of Young and Burgess.10 Patients with nondisplaced rami fractures were included if they did not require longer plate fixation. Age, sex, injury mechanism, Injury Severity Score (ISS), and use of posterior pelvic ring stabilization were recorded. Patients were excluded for bilateral injuries, associated acetabular fractures, displaced rami fractures, or if lost to follow-up before union. There were 96 patients (89 men and 7 women) with an average age of 46 years (range, 19–76 years). Patient and injury characteristics are listed in Table 1.
Two senior surgeons performed all of the operations at 2 separate level 1 trauma centers. Operative indications were the same for both centers, with iliosacral screws used only for type 3 injuries with widening and translational displacement of the posterior ring. Both centers used 6-hole 3.5-mm stainless steel plates through a rectus sparing approach. All plates were placed superiorly as described by Matta and Tornetta.11 One surgeon routinely removed the symphyseal cartilage, whereas the other did not. In the cartilage excision group, all cartilage was removed using an osteotome and curved curette. This was performed down to bone on both sides of the symphysis while protecting the urethra. The postoperative course was similar for both centers with partial weight-bearing for 6 weeks, then advancement to full weight-bearing as tolerated for APC type 2 injuries, and non-weight-bearing for 12 weeks for APC type 3 injuries.
All patients had a standard series of digital preoperative and postoperative pelvic radiographs including anteroposterior (AP), inlet, and outlet projections. Magnification was not controlled for using a reference ball; however, the PACS system adjusts for magnification in the area of interest, and all radiographs at both centers were measured the same way. These radiographs and the final follow-up radiographs were reviewed and form the basis of this radiographic study. The symphyseal space was measured at the superior and central aspects of the symphyseal bodies on AP and outlet projections by 2 observers using the distance annotation feature on the PACS system. These authors were not blinded to the treatment group but were not the surgeons involved in the care of these patients. The incidence of screw loosening/breakage and plate breakage was recorded from the final follow-up radiographs. Multiple screws loosening and/or breaking was considered one event. Plate breakage with screw loosening was considered one event in the combined calculation of construct failure.
Statistical analysis included 2-tailed t test for comparison of the continuous variables including initial patient demographics and radiographic displacement at all time points between the groups. The Fisher exact test was used for comparison of the discreet variables of implant failures, patient age, and revision surgery. The χ2 test was used for comparison of mechanism of injury. A P value of <0.05 was considered significant.
Fifty patients were treated at 1 institution with symphyseal cartilage removal, whereas 46 patients were treated at a second institution with cartilage retention. Motor vehicle accidents and pedestrians struck accounted for 67% of injuries. There was no statistical difference in mechanism between the groups (Table 1). The cartilage excision group was on average seven years younger (P = 0.015) and had a higher average ISS score (20.3 vs. 14; P = 0.01) than the cartilage retention group. Mean follow-up duration for the cartilage excision group was 17.3 months versus 13.2 months for the retention group. Table 1 shows patient demographics and injury characteristics.
There were 66 APC type 2 symphyseal injuries, all of whom had only anterior fixation by protocol. The initial preoperative displacement was not different between the groups (Table 2). As expected, the postoperative symphyseal space after cartilage excision was significantly less than if the cartilage was retained (Table 2). This difference was largest in the central symphysis as measured on the AP projection (2.4 mm vs. 6.0 mm). Statistical significance was maintained through union and was seen in all measurements made on the AP and outlet views. Table 2 shows radiographic position at final union as measured on the AP projection. This difference was largest in the superior symphysis (3.7 vs. 8.1 mm). The average change in alignment from immediately postoperatively to final union was 1.7 mm for the excision group and 2.1 mm for the retention group.
There were 30 APC type three symphyseal injuries in our series, all treated with anterior plating and posterior iliosacral screw fixation. There was no significant difference in preoperative displacement between the groups (Table 2). Again, immediately postoperatively, the symphyseal space with cartilage removal was less than if retained (Table 2). This difference was largest in the superior symphysis as measured on the AP projection (1.8 mm vs. 6.9 mm). As with the APC type 2 injuries, statistical significance was maintained through union and was seen in all measurements made on the AP and outlet views. The average change in alignment from immediately postoperatively to final union was 1.6 mm for the excision group and 2.1 mm for the retention group. Table 2 shows radiographic position at final union as measured on the AP projection. This difference was largest in the superior symphysis (3.6 vs. 8.9 mm).
There was no significant difference in the reduction quality of the symphysis between type 2 and type 3 injuries (P = 0.27); therefore, we report the overall differences in position for type 2 and 3 injuries together. Figure 1 illustrates AP projections preoperatively, immediately postoperatively, and a final union of an illustrative case of symphyseal cartilage excision.
The incidence of screw loosening/breakage, plate breakage, and combined implant problems was lower in the cartilage excision group. For APC type 2 injuries, plate breakage was the only measure that did not reach statistical significance (P = 0.16). Seven percent of the excision group and 19% of the retention group sustained a broken plate. Thirteen percent of the excision group versus 42% of the retention group had broken/loose screws (13%) (P = 0.015). Combined implant problems met statistical significance (P = 0.001) with 20% in the cartilage excision group and 61% in the cartilage retention group. For APC type 3 injuries, there were no significant differences in the incidence of screw loosening/breakage, plate breakage, or combined implant problems. There were no broken plates in the excision group versus one in the cartilage retention group (P = 0.40). Four of 20 patients had broken/loose screws (22%) in the excision group versus 1 of 10 in the retention group (8%) (P = 0.62).
The incidence of screw loosening/breakage, plate breakage, and combined implant problems were all significantly lower in the cartilage excision group when APC type 2 and type 3 injuries are reported together (Table 3). Eight patients in the excision group (16%) had screw loosening versus 16 patients in the retention group (35%) (P = 0.04). Only 2 patients in the excision group (4%) had plate breakage versus eight patients in the retention group (17%) (P = 0.05). Combined implant problems reached significance (P = 0.001). Figure 2 shows a case of hardware failure with resultant loss of reduction in the cartilage retention group.
As there was a difference in the demographic data with the retention group being slightly older but with lower ISS, we evaluated the effect of age and ISS on failure across both centers. Neither age nor ISS was statistically different for those who had implant failure or loosening (P = 0.17 and P = 0.96, respectively).
There were no reoperations in the cartilage excision group versus 4 in the cartilage retention group (P = 0.49). The indication for revision surgery was the same at both centers and performed only for catastrophic failure of fixation with redisplacement. Two of these revisions were performed for failure of anterior fixation. One was a patient with APC type 3 injury that required revision fixation 10 months postoperatively for progressive displacement with implant failure of both anterior and posterior fixation and continued pain. The second was a patient (body mass index, 43.1) with APC type 2 injury notable for gross noncompliance with weight-bearing restrictions, resulting in 2 revision surgeries for recurrent symphyseal widening and loss of fixation within 1 month postoperatively. This patient was ultimately treated in 6 weeks of traction after revision surgery and maintained acceptable alignment without evidence of implant failure at the final follow-up. A third patient required treatment for a sacral fracture nonunion, which was only possibly related to anterior motion. He had a body mass index of 37.5 and sustained an APC type 3 injury. He required sacral nonunion bone grafting 7 months postoperatively, followed by revision iliosacral screw fixation and bone grafting 11 months postoperatively. No revision of the symphyseal plate was required, although there was some widening, so its causal relationship is unclear. Finally, 1 patient had removal of symptomatic anterior heterotopic ossification 7 months postoperatively. At latest follow-up over four years after the patient's index surgery, there was no change in the original symphyseal fixation. This was not related to anterior instability.
Implant failure with recurrent widening of the symphyseal bodies is common after symphyseal reconstruction. To our knowledge, no data exist regarding excision of the symphyseal cartilage to gain better friction across the symphysis. We sought to evaluate the effect of symphyseal cartilage excision on final alignment and implant complications. Previous studies have reported high rates of implant complications after symphyseal plating that often results in recurrent symphyseal widening.2–8 Revision surgery for pain after implant failure may be required if this occurs.
Putnis et al reported implant failures and patient outcomes in a series of 49 patients with traumatic symphyseal diastasis treated with ORIF. Four- to eight-hole pelvic reconstruction plates were used. Radiographs were reviewed for screw breakage or backing out, change in plate position, or plate failure. Recurrent diastasis was also assessed. Functional outcomes using a validated SF-12 questionnaire and 6 questions specific to the pelvic injury were assessed at 1 year postoperatively. There were 15 patients with signs of movement (30.6%), 10 with loose screws, and 5 with broken screws. These findings were seen throughout different types of anterior and posterior fixation used, as well as when longer plates were used (25% of 6-hole plates). Of those with hardware failure, loss of reduction and recurrent symphyseal widening were seen in 6 patients (12%), 4 of which requiring revision surgery for anterior pelvic pain (8%). Functional outcome data were collected for 84% of the cohort. Patients with hardware failures had higher SF-12 physical and mental health scores; however, this did not reach statistical significance in this small series, leading the authors to conclude that hardware failure may not be an indication for revision fixation.2 In our series, no short plates were used, only 6-hole plates with 3 screws per side.
Other authors have drawn similar conclusions. Morris et al reviewed implant failures in 148 patients treated with a variety of plating methods; however, the 6-hole reconstruction plate was most commonly used. They reported implant breakage in 63 patients (43%), with plate breakage only in 14 (10%), plate and screws in 9, and screws only in 40 patients. Again, the rate of fixation failure was not found to be related to the type of plate used, type of injury, or posterior fixation. Revision surgery was performed in 3% for recurrent widening after implant complication. Four percent (6 patients) had loss of initial reduction, with 67% (2 of 6) of these required revision surgery.6
Sagi and Papp conducted a retrospective review of hardware failures in 92 operatively treated symphyseal disruptions with either a 2-hole 4.5-mm Limited Contact Dynamic Compression Plate (51 patients) or at least a 4-hole 4.5-mm pelvic symphyseal reconstruction plate (41 patients). In the 2-hole plate group, 33% had failure of fixation in the form of screw back out, all of which resulted in loss of initial postoperative reduction. Sixteen percent required reoperation for revision fixation or removal of implants. There was a high rate of pelvic malunion in this group at 57%. However, in the multihole plate group, 12% had failure of fixation in the form of plate breakage leading to loss of reduction and malunion. Additionally, 12% required reoperation for revision fixation or removal of painful implants. Overall, there was a 15% malunion rate for the multihole group.8 This is between the rates we report for the excision and the retention group.
In contrast, Giannoudis et al in a study on 74 patients reported a lower implant failure rate of 12%. Seven percent had loss of reduction; however, none required revision fixation.7 Similarly, Tornetta et al reported that 21% of rotationally unstable pelvic fractures treated with symphyseal plating had changes on their most recent radiograph including 4 broken plates compared with immediately postoperatively in their series of 29 patients in whom the cartilage was retained. However, none of these radiographic findings required reoperation.4 More recent studies have come to similar conclusions. In the series by Coolidge et al, they reported a 75% incidence of radiographic failure of fixation and 7% with broken plates. Ninety-three percent of patients with implant failure had recurrent symphyseal widening to a mean of 8.4 mm (range, 3–23 mm) at the final follow-up. Those with implant failure had a significant difference in recurrent widening compared with those without implant failure. However, only 1% required revision surgery. Implants used were either 4-hole or 6-hole pelvic reconstruction plates.3
Other studies that focus on outcomes after pelvic injury have found functional outcome measures of patients with operatively treated APC type injuries to be good to excellent compared with patients with posterior injuries; however, these studies did not focus on implant failures.12–14 There remains little consensus as to what degree of residual malalignment is acceptable in pelvic ring disruptions. It has been suggested that symphyseal fixation failure may represent relaxation of the implants as physiologic motion is restored; however, the clinical relevance of this motion remains unknown.3
In our series, all patients were treated with 6-hole reconstruction plates. One surgeon routinely removed the symphyseal cartilage during ORIF and the other did not. In all other aspects, the patients were managed the same way. Our results suggest that symphyseal cartilage excision led to substantially lower rates of implant failure while maintaining closer apposition of the symphyseal bodies through union. Although average age and ISS were different between the groups, neither factor was associated with higher combined implant failure rates. Previous studies suggest that outcomes are not as good if implant failure and resultant recurrent symphyseal widening occur with revision rates as high as 9%.2,3,6–8 We present one method to lower implant complications and pelvic malunion rates potentially leading to improved outcomes. The rate of revision surgery for malalignment was 3 of 46 (7%) for the retention group versus 0 of 50 (0%) for the excision group.
Plates are used to gain compression across the symphysis for stability through healing. We hypothesize that excision of the symphyseal cartilage would allow for greater friction with compression across bone-to-bone contact leading to improved stability, fewer implant complications, and possibly fewer revision surgeries. Although cartilage excision may result in increased strain at the center of the plate, this did not lead to higher failure rates of the plate.
The limitations of this study include its retrospective nature, small patient sample size, and nonuniform follow-up duration. Additionally, we do not report functional outcomes and therefore are unable to correlate the clinic implications of fewer implant complications and less recurrent symphyseal widening in the excision group.
We conclude that symphyseal cartilage excision results in lesser rates of recurrent symphyseal widening and fewer implant complications. Surgeons may elect to use this technique to avoid implant failure and maintain closer apposition of the symphyseal bodies through union.
1. Routt MLC, Nork SE, Mills WJ. High-energy pelvic ring disruptions. Orthop Clin North America. 2002;33:59–72.
2. Putnis SE, Pearce R, Wali UJ, et al.. Open reduction and internal fixation of a traumatic diastasis of the pubic symphysis. J Bone Joint Surg Br. 2011;93:78–84.
3. Collinge C, Archdeacon MT, Dulaney-Cripe E, et al.. Radiographic changes of implant failure after plating for pubic symphysis diastasis: an underappreciated reality? Clin Orthop Relat Res. 2012;470: 2148–2153.
4. Tornetta P, Dickson K, Matta J. Outcome of rotationally unstable pelvic ring injuries treated operatively. Clin Orthop Relat Res. 1996;329:147–151.
5. Raman R, Roberts CS, Pape HC, et al.. Implant retention and removal after internal fixation of the symphysis pubis. Injury. 2005;36:827–831.
6. Morris SA, Loveridge J, Smart DKA, et al.. Is fixation failure after plate fixation of the symphysis pubis clinically important? Clin Orthop Relat Res. 2012;470:2154–2160.
7. Giannoudis PV, Chalidis BE, Roberts CS. Internal fixation of traumatic diastasis of pubic symphysis: is plate removal essential? Arch Orthop Trauma Surg. 2008;128:325–331.
8. Sagi HC, Papp S. Comparative radiographic and clinical outcome of two-hole and multi-hole symphyseal plating. J Orthop Trauma. 2008;22:373–378.
9. Grimshaw CS, Bledsoe JG, Berton MR. Locked versus standard unlocked plating of the pubic symphysis: a cadaver biomechanical study. J Orthop Trauma. 2012;26:402–406.
10. Burgess AR, Eastridge BJ, Young JW, et al.. Pelvic ring disruptions: effective classification system and treatment protocols. J Trauma. 1990;30:848–856.
11. Matta J, Tornetta P. Internal fixation of unstable pelvic ring injuries. Clin Orthop Relat Res. 1996;329:129–140.
12. Van Loon P, Kuhn S, Hofmann A, et al.. Radiological analysis, operative management and functional outcome of open book pelvic lesions: a 13-year cohort study. Injury. 2011;42:1012–1019.
13. Pohlemann T, Gansslen A, Schellwald O, et al.. Outcome after pelvic ring injuries. Injury. 1996;27(suppl 2):B31–B38.
14. Dienstknecht T, Pfeifer R, Horst K, et al.. The long-term clinical outcome after pelvic ring injuries. Bone Joint J. 2013;95-B:548–553.