ORIGINAL ARTICLE

Results of Displaced Pediatric Tibial Spine Fractures: A Comparison Between Open, Arthroscopic, and Closed Management Eric W. Edmonds, MD,*w Eric D. Fornari, MD,z Jesse Dashe, MD,y Joanna H. Roocroft, MA,* Marissa M. King, BA,* and Andrew T. Pennock, MD*w

Background: Displaced tibial spine fractures are frequently treated with surgical reduction and fixation, but no comparison studies have been performed. This study was undertaken to compare fragment reduction and adverse outcomes between open arthrotomy [open reduction and internal fixation (ORIF)], arthroscopy [arthroscopic-assisted internal fixation (AAIF)], and closed management [closed management and casting (CMC)] of pediatric tibial spine fractures. Methods: A retrospective review of children treated for displaced tibial spine fractures from 2003 to 2011 was performed after categorizing into the 3 treatment groups. Demographics, mechanism of injury, radiographic measures (plain film and computed tomography), treatment, duration of immobilization and follow-up, final range of motion, and complications were recorded. Families were contacted to obtain long-term Lysholm scores, return to activity, pain, and satisfaction with treatment. Results: Seventy-six children (mean age, 12.4 y) met criteria with 29 ORIF, 28 AAIF, and 19 CMC. Radiographic measurements between x-ray and computed tomography scans found a mean error of 1 mm (SD = 1.33 mm; inter-class coefficient = 0.977, P < 0.001). Initial fracture displacement was similar between AAIF and ORIF, 10.3 ± 4.4 mm and 10.8 ± 3.9 mm; but, less in CMC group (5.3 ± 2.6 mm). The mean reduction amount was 8.6 ± 4.7, 9.1 ± 4.0, and 2.3 ± 2.6 mm, respectively. A Bonferroni post hoc analysis revealed a difference between surgical and nonoperative reduction (P < 0.001), but not between AAIF and ORIF (P = 0.9). Arthrofibrosis occurred with equal frequency in surgical cohorts (AAIF 12.5%, ORIF 11.1%), compared with none in the CMC group. Yet, the CMC group had a 16.7% risk for reoperation secondary to instability, loose bodies, or impingement. Twenty-four percent of each cohort was available (at mean 6.0 y) for interview with mean (median) Lysholm score: ORIF 97.4 (99), AAIF 95 (100), and CMC 86 (97.5), P = 0.35.

From the *Department of Orthopedic Surgery, Rady Children’s Hospital and Health Center; wDepartment of Orthopedic Surgery, University of California San Diego, San Diego, CA; zDepartment of Orthopedic Surgery, Montefiore Medical Center, Bronx, NY; and yDepartment of Orthopedic Surgery, Boston University School of Medicine, Boston, MA. The authors declare no conflicts of interest. Reprints: Eric W. Edmonds, MD, Department of Orthopedic Surgery, Rady Children’s Hospital and Health Center, 3030 Children’s Way, Suite 410, San Diego, CA 92123. E-mail: [email protected]. Copyright r 2014 Wolters Kluwer Health, Inc. All rights reserved.

J Pediatr Orthop



Conclusions: Open or arthroscopic treatment of displaced tibial spine fractures affords a better reduction than closed management, but with higher risk for arthrofibrosis. Closed management may be successful when displacement is 0.05, Table 1). However, 2 of the CMC patients reported their pain at a Z5; 1 child reported pain at a 5 which they felt increased to an 8 during activity (this patient also reported locking and instability), and another child reported pain at a 9 (Table 2). One child in the ORIF group reported pain at a 4, which they specified was random and occurred during exertion. No significant differences were observed with respect to Lysholm score or treatment satisfaction (on a scale: 0 to 10, 10 being completely satisfied), P > 0.05 (Table 1). No families reported requiring further operative intervention.

DISCUSSION

FIGURE 1. An 11-year-old boy who fell while skateboarding: plain radiographic measurement of displacement (A) and computed tomographic measure of maximal displacement (B).

delay for those with arthrofibrosis in the open cohort was 8 days (range: 3 to 14 d); basically, the same mean to the remaining children without evidence for arthrofibrosis. Although free of arthrofibrosis, the CMC cohort resulted in a 16.7% risk for subsequent operation secondary to instability, loose bodies, or impingement. These surgeries included subsequent conversion to operative fixation of the fragment, ACL reconstruction, and excision of the impinging fragment or loose body. The children requiring a future operation in the initial CMC cohort had a mean displacement of 6.7 mm (range: 4.2 to 10.9 mm). All, but one, of the children with arthrofibrosis underwent a secondary sedation event for either manipulation under anesthesia or arthroscopic lysis of adhesions. Moreover, one Copyright

r

2014 Wolters Kluwer Health, Inc. All rights reserved.

Treatment options for displaced tibial spine fractures include closed reduction and cast immobilization, open reduction and internal fixation, and arthroscopicassisted reduction, and internal fixation. To the best of our knowledge, our study is the first to compare the results of these 3 treatment modalities in the management of type II, III, and IV pediatric tibial spine fractures. As might be expected, greater amount of reduction was obtained for those treated operatively compared with closed reduction and immobilization; however, there was no difference between the 2 surgical cohorts (Fig. 2). Furthermore, our CMC cohort had a higher reoperation rate for late instability, loose bodies, and impingement; which is consistent with previous literature demonstrating TABLE 1. Results of the Outcome Interviews (n = 18) Pain (0-10) Lysholm Treatment satisfaction (0-10)

ORIF

AAIF

CMC

P

0.7 (0) 97.4 (99) 9.9 (10)

0.2 (0) 95 (100) 9.6 (10)

2.7 (1) 86 (97.5) 9 (9.25)

0.32 0.35 0.25

Findings are reported as mean (median) due to the low sample size. Kruskal Wallis nonparametric test was performed on the 3 measures. AAIF indicates arthroscopic-assisted internal fixation; CMC, closed management and casting; ORIF, open reduction and internal fixation.

www.pedorthopaedics.com |

Copyright r 2015 Wolters Kluwer Health, Inc. All rights reserved.

653

J Pediatr Orthop

Edmonds et al



Volume 35, Number 7, October/November 2015

TABLE 2. Detailed Outcome Information on the 18 Patients Completing the Follow-up Interview Patient #

Initial Treatment Age at Displacement Group Injury (y) (mm)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

ORIF ORIF ORIF ORIF ORIF ORIF ORIF AAIF AAIF AAIF AAIF AAIF CMC CMC CMC CMC CMC CMC

13 12 11 9 7 12 13 17 15 11 9 14 12 3 12 12 16 13

17.1 9.2 7.5 13.6 8.3 13.3 5.6 11.4 13.5 8.3 — 6.8 — 4.3 10 — 1.9 9

Age at Outcome (y) 17 22 20 14 16 21 19 16 18 21 20 17 17 8 17 22 20 19

Years of Lysholm Pain Follow-up Score Score 3 9 8 5 9 8 5 3 2 10 11 3 5 5 5 10 4 6

100 80 100 100 100 100 100 81 100 99 95 100 67 100 100 100 52 95

4 1 0 0 0 0 0 1 0 0 0 0 5 0 0 0 9 2

Treatment Return to Previous Further Injury Satisfaction Level of Activity or Surgery (1-10) Yes No Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes

No No No No No No No No No No No No No No No No No No

9.5 10 8.5 10 10 10 10 10 10 10 9.5 10 10 10 8 10 7.5 8.5

Patients #11, 13, and 16 had incomplete initial radiographs, but chart documentation indicated #13 was nondisplaced, and #16 after casting computed tomography measured 3 mm displacement. AAIF indicates arthroscopic-assisted internal fixation; CMC, closed management and casting; ORIF, open reduction and internal fixation.

that nonsurgically treated fractures healing with significant residual displacement result in lack of extension due to impingement within the notch.8 Persistent objective laxity is well documented following tibial spine fractures, even though it does not always correlate with subjective instability. Kocher et al27 reported persistent anterior laxity, with an abnormal Lachman examination in 5 of 6 patients, abnormal pivot shift in 2 of 6, and increased KT-1000 laxity in 4 of 6 patients, despite excellent functional scores in all 6 patients. These results are consistent with those of Baxter and Wiley3,33 who reported excellent functional results without instability in 17 patients despite a positive Lachman examination in 51% of patients and increased mean instrumented laxity of 3.5 mm. Furthermore, Smith31 found subjective instability in only 2 of 13 patients following ORIF of type III fractures despite a positive Lachman examination in 87% of patients. Willis et al32 similarly reported excellent clinical results in 50 patients despite a positive Lachman examination in 64% and instrumented knee laxity of 3.5 mm for type II fractures and 4.5 mm for type III fractures. This objective knee laxity is thought to be related to interstitial injury of the ACL with elongation at the time of tibial spine fracture.27 The main complication encountered in both our surgical cohorts was arthrofibrosis. Although loss of extension is well documented with this injury, recent literature has specifically focused on arthrofibrosis following operative management. Vander Have et al29 reported on a series of 32 patients with knee stiffness following surgical fixation. Of these, 24 went on to require additional procedures and 3 sustained intraoperative distal femur fractures and subsequent growth arrest. Within their cohort there was a 10% rate of arthrofibrosis.29 Patel et al28 reported that 7 of 40 patients required additional surgeries for

654 | www.pedorthopaedics.com

arthrofibrosis at a mean of 3 months after initial fracture treatment. In their cohort, those who started range of motion within 4 weeks of treatment were 12 times less likely to develop arthrofibrosis. Our rates of arthrofibrosis are consistent with these previous reports. However, the duration of immobilization was not different between our surgical and nonoperative cohorts suggesting that surgery itself, rather than duration of immobilization, may be the greater risk for arthrofibrosis. The current study found that 32% of the fracture reductions were being impeded by an entrapped meniscus or intermeniscal ligament at the time of surgery. Meniscal entrapment has been well documented in association with tibial spine fractures. Kocher et al34 found meniscal entrapment preventing reduction in 26% of type II fractures and 65% of type III fractures. This is especially important to consider in type II fractures that do not reduce in extension. Burstein et al35 and Chandler and Miller36 also reported cases of meniscal incarceration blocking reduction of types II and III tibial spine fractures. Finally, Mah et al17 reported medial meniscal entrapment preventing reduction in 9 of 10 patients with type III fractures. Preoperatively, all patients in our series had both x-rays and CT scans; but, our series found that the measures of fracture displacement were nearly identical between plain film and CT, calling into question the utility of advanced imaging to determine displacement. However, it does not exclude the potential benefit of identifying possible articular extension of the fragment, soft-tissue entrapment, or unrecognized fracture comminution. Griffith et al10 previously evaluated the benefit of CT versus plain radiographs and found that preoperative CT scans offered improved understanding of avulsed fragment shape and extent of tibial plateau involvement. Moreover, Yacobian et al13 found Copyright

r

2014 Wolters Kluwer Health, Inc. All rights reserved.

Copyright r 2015 Wolters Kluwer Health, Inc. All rights reserved.

J Pediatr Orthop



Volume 35, Number 7, October/November 2015

Results of Displaced Pediatric Tibial Spine Fractures

FIGURE 2. Comparison of preoperative (left) and final postoperative (right) images. A, Arthroscopically assisted reduction and fixation of a 15-year-old male skateboarder with 16 months radiographic follow-up. B, Open reduction and internal fixation of an 11-year-old female dirt biker with 3 months of radiographic follow-up.

MRI useful for determining the degree of displacement and comminution of tibial eminence fractures, thereby eliminating the need for CT and the radiation associated with this diagnostic modality. The treatment of displaced tibial spine fractures via closed management and casting achieved less fracture reduction compared with the surgically treated children. However, this method of treatment had a much lower risk of arthrofibrosis; and yet, a much higher risk for subsequent surgery for laxity and impingement. Those that required a surgery after initial nonoperative management had a mean displacement of 6.7 mm, suggesting that perhaps a safe cut-off for nonoperative management would be r5 mm. We feel that this amount of displacement has a low risk for subsequent surgery, while simultaneously reducing the risk for arthrofibrosis that comes with surgical management. Limitations of this study include its retrospective design and inability to contact all the eligible children for Copyright

r

2014 Wolters Kluwer Health, Inc. All rights reserved.

long-term follow-up. Some children were 10 years postoperative, and this is a highly migratory population given that current age would place them already graduated from college and no longer with viable telephone numbers on file. However, as the study is predominately a radiographic outcome study, and all enrolled children were followed to fracture union, many of the limitations of retrospective design are reduced. Moreover, another limitation inherent to retrospective design includes the inability to control for the nuances of treatment, including: multiple surgeons, angle of knee flexion with casting, weight-bearing status, or implants of fixation. Another potential limitation is our omission to obtain objective laxity measures. However, as discussed earlier, multiple authors have demonstrated a lack in correlation between functional outcomes and objective laxity measures downplaying the importance of residual laxity. Finally, there was a significant difference in the duration of clinical follow-up between the surgical cohorts and the nonoperative cohort; however, after applying www.pedorthopaedics.com |

Copyright r 2015 Wolters Kluwer Health, Inc. All rights reserved.

655

J Pediatr Orthop

Edmonds et al

exclusion criteria, all included patients had radiographic evidence of fracture union. Thus, the implication is that the surgical groups required more time to rehabilitate after treatment than the nonoperative group. In conclusion, our series represents one of the largest cohorts of children with displaced tibial spine fractures managed at a single institution. Minimally displaced type II fractures (< 5 mm elevation) can be successfully managed with closed reduction and casting. However, when evaluating the initial displacement (not the postreduction displacement), one should have a low threshold for performing operative fixation in those with >5 mm fragment displacement, considering

Results of Displaced Pediatric Tibial Spine Fractures: A Comparison Between Open, Arthroscopic, and Closed Management.

Displaced tibial spine fractures are frequently treated with surgical reduction and fixation, but no comparison studies have been performed. This stud...
591KB Sizes 0 Downloads 8 Views