Sacral fractures are common in the setting of pelvic ring injury. The most common type is a unilateral sacral injury with anterior impaction of the sacrum, commonly known as a lateral compression type 1 (LC-1) injury. Historically, these injuries have been treated nonoperatively with the expectation of union in an acceptable position.1,2 However, others have adopted the position that sacral fractures should be treated operatively. At this time, the operative indications are unclear even in undisplaced fractures. One of the drivers of operative management is the belief that fixation will diminish early pain, potentially leading to faster recovery.3
To better assess the need for surgery in patients with unilateral sacral fractures, we designed a multicenter prospective trial funded by the Orthopaedic Trauma Association to evaluate the outcomes of patients treated operatively and nonoperatively. We previously reported that operative treatment is common even in patients with no displacement preoperatively.4 The purpose of this report is to compare the pain experienced from injury to union (12 weeks) by patients with undisplaced or minimally displaced unilateral sacral fractures treated operatively versus those treated nonoperatively.
PATIENTS AND METHODS
Over a 7-year period, we offered enrolment to all adult skeletally mature patients who were fully ambulatory without assistive devices with unilateral sacral fractures in 16 centers. Exclusion criteria were anteroposterior (AP) compression injuries as demonstrated by symphyseal dissociation, displaced acetabulum fractures, fractures that crossed the mid-line (zone 3), pregnant patients, prisoners, nonambulatory patients, and those who would not be able to follow-up. Surgeons were not asked to change their method of treatment in any way including their indications for surgery. Surgery was based on the radiographic analysis of the injury rather than other patient factors.
In addition to standard demographic information, all fractures were evaluated radiographically on plain films and computed tomography. Fractures were classified by the method of Young and Burgess.5 Plain films consisted of supine AP, inlet, and outlet views. Computed tomography was by protocol at each center. A PowerPoint presentation was sent to all centers to instruct them regarding the specific method to make all measurements. These included the displacement in millimeters (mm) of the sacrum, upper ilium, and ischium on the AP and outlet views, and the sacrum and the posterior ilium on the inlet views.
For the purpose of this report, fractures that demonstrated less than 5 mm of translational displacement (excluding impaction) of any posterior structure on the AP and inlet views were included, as these were considered minimally or undisplaced. The differences in ring width (as an absolute value) were measured for all patients on the inlet view as the distance from the mid-line to the quadrilateral surface on each side, but these measurements were not used as an inclusion or exclusion, only as a comparison of asymmetry. Pain was assessed using a standard Visual Analog Scale (VAS) score of 0–10.6–8 Pain scores “over the last day” in the anterior pelvis and the posterior pelvis were documented at baseline (before injury per recall). Pain was then documented within 24 hours after surgery for those who were treated surgically and for the first 24 hours of admission for those treated nonoperatively. Pain scores were obtained at 1, 3, 6, and 12 weeks after treatment or after injury. Some patients were reached by telephone and then by mail if they were not able to return to the outpatient clinic for follow-up at the desired interval.
We compared the VAS pain at each time point for all patients with data at that visit using 1-way analysis of variance. Comparisons of operative with nonoperative patients were made using the Student t test for continuous variables and the Fisher exact test for categorical values. SPSS software version 24 (IBM, Armonk, NY) was used. Significance was set at P < 0.05 for all analyses.
We enrolled 334 patients with sacral fractures of whom 194 met the inclusion criteria for displacement and had adequate radiography. Mean age was 38.7 years, mean Injury Severity Score (ISS) was 14.5%, and 61% of patients were women (Table 1). The average body mass index (BMI) was 26.2. The most common mechanisms of injury were motor vehicle collision (56%) followed by fall from a height (19%). Ninety-nine percent of fractures were classified as LC injuries. One hundred forty-four patients were treated nonoperatively (74%) and 50 operatively (26%). For the operatively treated patients, 19 had posterior only fixation, 2 had anterior only fixation, and 29 had both posterior and anterior fixation. All patients with posterior fixation had at least 1 SI screw. Owing to the small number of surgical patients and the number of centers (and surgeons per center), no realistic analysis of tendencies by individual surgeon could be made. There were no differences in age, sex, or BMI between the groups. Mechanisms of injury were similar between the operative and nonoperative groups, although operative patients were less likely to be in a motor vehicle collision (44% vs. 60%) and more likely to have had a motorcycle crash (10% vs. 2.8%). ISS indicated multiple injuries were common in both groups (17.1 operative vs. 13.6; P = 0.13). Patients with sacrum fractures in zone 1 were more likely to be treated nonoperatively (63% vs. 40% operative, P = 0.004) than fractures in zone 2. See Table 1. There was no difference in ring width asymmetry in the operative versus the nonoperative group (3.3 vs. 3.1 mm P = 0.8), and 72% had less than 5-mm difference. Nonoperatively treated patients were allowed to mobilize and weight-bear fully as comfort dictated. Operatively treated patients were treated by the individual surgeons' protocols.
Preinjury VAS scores for pain in the pelvic ring were no different between the treatment groups. However, nonoperative patients reported a mean VAS 2.7 points higher in the posterior pelvis (P = 0.01) and 1.9 points higher in the anterior pelvis (P = 0.11) at 24 hours after treatment or after injury when compared with patients treated operatively. Measurements obtained 6 weeks later showed slightly higher VAS for pain in posterior pelvis for nonoperative patients versus those treated surgically (5.2 vs. 3.5, P = 0.033), with no difference in the anterior ring. VAS scores in both treatment groups continued to improve over time, but did not return to baseline levels by the time of union. Twelve weeks after injury, the mean VAS for pain was 1.1 point more in the posterior pelvis for nonoperative patients versus those treated surgically (4.0 vs. 2.9, P = 0.019), along with a slight difference in the VAS for anterior pain (3.2 vs. 2.3, P = 0.035). No symptomatic nonunions were identified, and no patient had displacement of their sacrum on the AP radiograph of more than 4 mm. Data are shown in Figures 1 and 2.
The treatment of sacral fractures is controversial with some surgeons treating those with small displacements nonoperatively and others treating undisplaced fractures with surgery.1,9–20 In a group of 38 patients with unstable pelvic fractures, Barei et al3 demonstrated a significant decrease in both VAS pain scores and narcotic use after surgical stabilization. Hagan et al in a single-center evaluation retrospectively reviewed patients with LC (LC-1) and LC-2 injuries with an attempt to match injuries in the operative and nonoperative groups (based on surgeon preference). They found little difference in narcotic use or pain when comparing the operative and nonoperative cohorts. However, patients were included in their study if they had less than 10 mm of displacement. In addition, pain scores in this trial were not specific to the pelvic injury or location (anterior vs. posterior).
In an attempt to evaluate what we considered to be the most amenable group to nonoperative treatment, we evaluated only in patients with undisplaced fractures or less than 5 mm of displacement on initial radiographs. If there were an advantage to operative treatment in this group, then patients with more displaced fractures would likely also see a benefit.3 Multiple trials have shown that minimally displaced fractures have little to no propensity to displace over time; so, the only indication for operative intervention would be for pain relief, which could potentially diminish pain medication requirements or aid in early mobilization.1,2,10
This trial reports on a subset of a prospective Orthopaedic Trauma Association–funded study of unilateral sacral fractures. Treatment was based on surgeon preference because most surgeons were unwilling to randomize to operative versus nonoperative management. The analysis of early pain and of late outcomes in the overlap group of patients with fracture displacements, which were treated both operatively and nonoperatively by different surgeons, was a planned analysis. In reviewing the data on displacement, we found that many patients had no displacement, and approximately one-quarter (26%) were treated operatively. This gave us an almost ideal group to evaluate for outcomes because the indication for surgery in this group was not for fracture reduction.
We found differences in anterior or posterior pain between treatment groups based on the 10-point VAS at various time points. At 6 weeks, there was no difference between the 2 groups in anterior pain, while a 1.7-point difference was noted posteriorly, favoring the operative group. There was no difference in union at 3 months, but a 1-point reduction in posterior pain was noted in the operative group. While statistically significant, the minimally important clinical difference (MCID) remains uncertain.21 For example, the MCID on the VAS for pain due to osteoarthritis of the knee has been shown to be a reduction of 2.5 for surgical or rehabilitative therapies.22 Despite this, surgical treatment may provide a minor improvement in pain relief, and pain continues to improve over time for both treatment groups.
Weaknesses of this study include lack of measurements of other objective pain scores or consumption of pain medication. However, the VAS has been shown to be reliable for patients of various ages, and with various educational and literacy levels.7,8,23 The VAS is also considered sensitive for measuring pain and has been shown to correlate with other objective measurements of pain intensity.24 We also did not account for the timing of consumption of pain medication or of physical activity of patients relative to obtaining the VAS. It is likely that during the study, VAS scores were recorded shortly after administration of medication for some patients, and much later for others; however, this variability, which may result in a larger range for VAS responses, would be expected to be present with equal frequency in the operative and nonoperative groups. We also did not account for the presence of previous chronic pain, baseline prescription pain medication, or recreational drug consumption.5,26 Again, these patient factors may have been present with similar distribution among the 2 study groups. Finally, the initial VAS scores for the operative group were recorded immediately postoperatively, and so, we cannot determine whether they had more pain than those treated nonoperatively. However, the decision to operate was made based on the initial radiographs, and there was a large variation in the nonoperative group that supports the supposition surgery was not based on initial presenting pain. Strengths of this study include prospective collection of data on a homogeneous group of patients at specified time points after injury at multiple high-volume trauma centers throughout North America.
Based on this trial, there was limited early pain relief within 3 months after injury for operatively treated minimally or undisplaced sacrum fractures. It is unknown whether the difference in the VAS represents clinical relevance; however, the difference is below the MCID for the VAS for other orthopaedic conditions. It is unclear whether these findings can be extrapolated to patients with displaced fractures or to those with other posterior pelvic ring injuries.
The study was coordinated at Boston University Medical Center, Boston, Massachusetts. Participating centers included Harborview Medical Center, Seattle, Washington; MetroHealth Medical Center, Cleveland, Ohio; Orthopaedic Associates of Michigan, Grand Rapids, Michigan; Eskenazi Hospital, Indianapolis, Indiana; Boston University Medical Center, Boston, Massachusetts; University of Oklahoma, Oklahoma City, Oklahoma; Indiana University Methodist Hospital, Indianapolis, Indiana; Oregon Health Sciences University, Portland, Oregon; University of Alabama, Birmingham, Alabama; Wake Forest University, Winston-Salem, North Carolina; Hennepin County Medical Center, Minneapolis, Minnesota; University of Mississippi Medical Center, Jackson, Mississippi; University of Utah, Salt Lake City, Utah; University of Texas Health Science Center, Houston, Texas; Barnes Jewish Hospital, Washington University School of Medicine, St. Louis, Missouri; Halifax Infirmary, Halifax, Nova Scotia; and Zuckerberg San Francisco General Hospital, San Francisco, California.
1. Bruce B, Reilly M, Sims S. OTA highlight paper predicting future displacement of nonoperatively managed lateral compression sacral fractures: can it be done? J Orthop Trauma. 2011;25:523–527.
2. Sembler Soles GL, Lien J, Tornetta P III. Nonoperative immediate weightbearing of minimally displaced lateral compression sacral fractures does not result in displacement. J Orthop Trauma. 2012;26:563–567.
3. Barei DP, Shafer BL, Beingessner DM, et al. The impact of open reduction internal fixation on acute pain management in unstable pelvic ring injuries. J Trauma. 2010;68:949–953.
4. Tornetta P III, Agel J, Miller A, et al. Does operative intervention provide early pain relief for patients with undisplaced sacral fractures? Podium presentation at the Annual Meeting of the Orthopaedic Trauma Association; October 2016; Baltimore, MD.
5. Burgess AR, Eastridge BJ, Young JW, et al. Pelvic ring disruptions: effective classification system and treatment protocols. J Trauma. 1990;30:848–856.
6. Burkhardt CS, Jones KD. Adult measures of pain: The McGill Pain Questionnaire (MPQ), Rheumatoid Arthritis Pain Scale (RAPS), Short-form McGill Pain Questionnaire (SF-MPQ), Verbal Descriptive Scale (VDS), Visual Analog Scale (VAS), and West haven-Yale Multidisciplinary Pain Inventory (WHYMPI). Arthritis Rheum. 2003;49:S96–S104.
7. Hawker GA, Mian S, Kendzerska T, et al. Measures of adult pain. Arthritis Care Res. 2011;63:S240–S252.
8. Singh AP. Visual analog scale for pain, bone spine [bone and spine website]. Available at: www.boneandspine.com/visual-analog-scale-for-pain
. Accessed February 10, 2018.
9. Gaski GE, Manson TT, Castillo RC, et al. Nonoperative treatment of intermediate severity lateral compression type 1 pelvic ring injuries with minimally displaced complete sacral fracture. J Orthop Trauma. 2014;28:674–680.
10. Hagen J, Castillo R, Dubina A, et al. Does surgical stabilization of lateral compression-type pelvic ring fractures decrease patients' pain, reduce narcotic use, and improve mobilization? Clin Orthop Relat Res. 2016;474:1422–1429.
11. Beckmann JT, Presson AP, Curtis SH, et al. Operative agreement on lateral compression-1 pelvis fractures. a survey of 111 OTA members. J Orthop Trauma. 2014;28:681–685.
12. Bodzay T, Sztrinkai G, Pajor S, et al. Does surgically fixation of pubic fracture increase the stability of the operated posterior pelvis? Eklem Hast Cerrahisi. 2014;25:91–95.
13. Emohare O, Slinkard N, Lafferty P, et al. The effect of early operative stabilization on late displacement of zone I and II sacral fractures. Injury. 2013;44:199–202.
14. Griffin DR, Starr AJ, Reinert CM, et al. Vertically unstable pelvic fractures fixed with percutaneous iliosacral screws: does posterior injury pattern predict fixation failure? J Orthop Trauma. 2003;17:399–405.
15. Jones CB. Posterior pelvic ring injuries: when to perform open reduction and internal fixation. Instr Course Lect. 2012;61:27–38.
16. Lefaivre KA, Padalecki JR, Starr AJ. What constitutes a Young and Burgess lateral compression-I (OTA 61-B2) pelvic ring disruption? A description of computed tomography-based fracture anatomy and associated injuries. J Orthop Trauma. 2009;23:16–21.
17. Matta JM, Tornetta P III. Internal fixation of unstable pelvic ring injuries. Clin Orthop Relat Res. 1996;329:129–140.
18. Routt ML Jr, Simonian PT. Closed reduction and percutaneous skeletal fixation of sacral fractures. Clin Orthop Relat Res. 1996;329:121–128.
19. Sagi HC, Militano U, Caron T, et al. A comprehensive analysis with minimum 1-year follow-up of vertically unstable transforaminal sacral fractures treated with triangular osteosynthesis. J Orthop Trauma. 2009;23:313–319; discussion 9-21.
20. Tötterman A, Glott T, Søberg HL, et al. Pelvic trauma with displaced sacral fractures: functional outcome at one year. Spine. 2007;32:1437–1443.
21. Bodian CA, Freedman G, Hossain S, et al. The visual analog scale for pain: clinical significance in postoperative patients. Anesthesiology. 2001;95:1356–1361.
22. Katz PK, Paillard FC, Ekman. Determining the clinical importance of treatment benefits for interventions for painful orthopaedic conditions. J Orthop Surg Res. 2015;10:24.
23. Breivik H, Borchgrevink PC, Allen SM, et al. Assessment of pain. Br J Anaesth. 2008;101:17–24.
24. Haefeli M, Elfering A. Pain assessment. Eur Spine J. 2006;15(suppl 1):S17–S24.
25. Gabrielle Page M, Fortier M, Ware MA, et al. As if one pain problem was not enough: prevalence and patterns of coexisting chronic pain conditions and their impact on treatment outcomes. J Pain Res. 2018;11:237–254.
26. Denis F, Davis S, Comfort T. Sacral fractures: an important problem. Retrospective analysis of 236 cases. Clin Orthop Relat Res. 1988;227:67–81.