Closed Reduction in a “Hyperextended Supine Position” with Percutaneous Transsacral-Transiliac and Iliosacral Screw Fixation for Denis Zone III Sacral Fractures

Background Herein, we demonstrate the clinical results of closed reduction in a hyperextended supine position with transsacral-transiliac (TSTI) and iliosacral (IS) screw fixations for Denis zone III sacral fractures. Patients and Methods Sixteen consecutive patients with Denis zone III sacral fractures who were treated between January 2009 and September 2016 were evaluated. All patients were treated using percutaneous TSTI/IS screw fixation during closed reduction performed with patients placed in a hyperextended supine position with body manipulation. The clinical and radiological results were evaluated, and the neurological outcomes were retrospectively assessed using Gibbon's classification system. The clinical outcomes were evaluated using the German Multicenter Study Group Pelvic Outcome Scale (POS). Results The sacral kyphotic angle was reduced by 18.06° ± 15.26° (mean kyphotic angle: pre-OP, 39.44° ± 20.56°; post-OP, 21.38° ± 7.39°), and fracture translation was reduced by 5.93 ± 4.95 mm (mean fracture translation: pre-OP, 8.69 ± 8.03 mm; post-OP 2.75 ± 3.97 mm). The mean initial Gibbon's score was 3.00 ± 1.32. Among 15 patients with a follow-up duration of over 12 months, the mean reduction loss in the sacral kyphotic angle was 5.87° ± 10.40° and was 1.00 ± 3.00 mm for the fracture translation. The final Gibbon's score was 1.80 ± 1.21, and 73.3% of patients had good results based on the POS score. Conclusions Although closed reduction in a hyperextended supine position with percutaneous posterior screw fixation is associated with some surgical limitations and technical difficulties, the procedure is minimally invasive and highly effective for stabilizing Denis zone III sacral fractures.


Introduction
Denis zone III sacral fractures are generally due to highenergy traumas such as falls from heights, traffic accidents, and crush injuries, with most of these fractures occurring in patients with polytrauma [1]. Zone III sacral fractures are relatively rare injuries, reportedly accounting for only 3-5% of all sacral fractures [2][3][4]. Importantly, these fractures are associated with a high rate of neurologic injury, including sensory and motor deficits in the lower legs; saddle anesthesia; and bowel, bladder, and sexual dysfunctions [2].
Historically, the standard treatment for Denis zone III sacral fractures did not involve surgery or was limited to sacral laminectomy [1,5]. However, given the advances in internal fixation techniques, cases with minimal kyphosis and no neurological deficits can be treated using several commonly performed surgical procedures such as spinopelvic fixation [6], posterior plate fixation [7], and iliosacral screw fixation [8]. However, no clear guidelines currently exist regarding appropriate treatment strategies and indications for Denis zone III sacral fractures. Spinopelvic and posterior plate fixation have shown good reduction and neurological recovery but are associated with long operating times. Moreover, the surgery is performed in the prone position, which may be particularly disadvantageous for polytrauma patients.

Advances in Orthopedics
In addition, the use of posterior plates and other spinal instrumentations may cause soft tissue damage and lead to wound complications. To minimize the risk of such complications, minimally invasive approaches have been increasingly used for these fractures. In recent reports, minimally invasive posterior screw fixations for pelvic ring fractures, including transsacral-transiliac (TSTI) and iliosacral (IS) screw fixations, have been reported to have a number of clinical and biomechanical advantages [9][10][11][12][13].
The purpose of the present study was to review our experiences and demonstrate the clinical results of closed reduction in a hyperextended supine position with percutaneous TSTI/IS screw fixation methods for Denis zone III sacral fractures.

Patients and Methods
Patients with Denis zone III sacral fractures classified as AO/OTA type C pelvic injuries, treated between January 2009 and September 2016, were selected from our trauma database. Patients who had undergone surgical treatment and had been followed up for a minimum of 12 months were selected for analysis. The patients' medical records were reviewed retrospectively, and the fractures were classified by the sacral level according to the method of Roy-Camille (as modified by Strange-Vognsen and Lebech) and the fracture morphology [1,14,15].
The following injury data were collected: demographics, mechanism of injury, associated injury, and the ISS. An initial clinical neurological examination was performed if the patient's condition permitted and was graded according to the classification system described by Gibbons et al [5,16]. Preoperative imaging consisted of pelvic anterior-posterior radiographs and multislice CT scans. Collected postoperative data included the results from clinical and radiological assessments. Radiographic assessments of the sacral kyphotic angle, fracture translation, implant position, and decompression were performed using standard radiography of the pelvis and multislice CT [17,18]. Additionally, multidisciplinary followup examinations were performed by an orthopedic surgeon (H.I.).

Reduction and Operative Techniques.
During the 8-year study period, a standardized operative technique using closed reduction and posterior pelvic ring fixations (TSTI and/or IS screw fixation) was performed. In patients with combined sacral fractures and anterior pelvic ring injuries, staged and/or one-stage reconstruction was performed. Injuries of the anterior part of the pelvic ring were treated using an anterior extraperitoneal or percutaneous approach. Posterior pelvic ring fixation was performed through the percutaneous approach.

Operative
Procedure. For posterior screw fixation, the patients were positioned supine on a radiolucent table and in hyperextension with handmade pillows placed vertical to the sacral fracture line (Figure 1(a)). Under fluoroscopic guidance, closed fracture reduction was performed. When the fracture reduction quality was insufficient, additional pillows were placed and/or manipulation of the patient's body was performed (Figure 1(b)). The manipulation maneuver involved putting a tow in the longitudinal direction while pushing the trunk and lower limbs to the table (Figure 1(c)). After closed reduction, percutaneous screw insertion was performed under fluoroscopic guidance (Figure 1(d)). In principle, two or more TSTI screws were inserted; when this was not possible, the TITS screws were set in combination with an IS screw. Screws utilized at our institute included 6.5 mm (diameter) cannulated cancellous screws provided by Depuy-Synthes, Inc. (partial thread; maximum length: 120 mm) and Meira, Inc. (partial and full thread; maximum length: 150 mm). Direct sacral decompression was not initially performed.

Postoperative
Care. The patient was allowed to sit with the torso upright starting on postoperative day 1. Nonweight bearing activities were allowed 4 weeks after definitive surgery. Weight-bearing activities began 5 weeks after definitive surgery.
After definitive pelvic ring fixation, a follow-up CT was performed within 1 week to indirectly evaluate the decompression quality. If a remnant fracture fragment in the sacral canal and/or poor indirect decompression quality was found, a secondary direct sacral decompression was considered. One orthopedic trauma surgeon (H.I.) performed all operative procedures. Early and late complications associated with the surgical treatment were recorded.

Radiological Evaluation.
The sacral kyphotic angulation of the Denis zone III sacral fracture was measured from sagittal CT reformations by measuring the angle between the posterior sacral cortices, superior and inferior to the level of the transverse fracture. Fracture translation was also measured from the sagittal CT reformations by measuring the displacement of the anterior cortex of the sacrum above and below the transverse fracture ( Figure 2). All measurements involving the preoperative, postoperative, and final radiographs, as well as the CT images, were performed by the first author.

Outcome Evaluation.
Lower extremity sensory and motor function and rectal examinations were performed pre-and postoperatively to identify injuries to the lower lumbosacral plexus. Neurological deficits from cauda equina injuries were classified according to the method of Gibbons et al [5]. Improvements in neurological function at the final follow-up were similarly assessed.
The clinical outcomes were evaluated at the final followup using the clinical criteria of the German Multicenter Study Group Pelvis Outcome Scale (POS) [19]. These clinical criteria (pain, functional impairment, persistent neurological and urological impairments, and bowel dysfunction) are based on the clinical results and range from 1 to 4 points; a POS score of 3-4 points is considered to be a good outcome, whereas a score of 1-2 points indicates a poor outcome. The left image was taken before reduction maneuver and the right image was taken after the maneuvers. If good reduction was obtained, screw insertion was done in the same position. (e) Intraoperative fluoroscopic images pre-and postreduction maneuver. In this case, good reduction was obtained.

Statically Analysis. Data was enrolled through Microsoft
Excel 2016 (Microsoft, Redmond, WA), followed by a statistical analysis using IBM SPSS Statistics, version 19 (SPSS, Chicago, IL). Data are given in terms of arithmetic mean and standard deviation. The initial, postoperative, and final follow-up data were analyzed using paired t-test. Data differences were considered significant for values of p < 0.05. (Table 1). Sixteen patients (6 men, 10 women) with Denis zone III sacral fractures classified as AO/OTA type C pelvic injuries were identified. Table 1 presents the patient demographics, mechanism of injury, ISS, fracture patterns, initial Gibbons' grade, and operative procedures. At the time of injury, the mean age of the patients was 29.50 ± 11.12 years (range: 16-50 years). All patients sustained high-energy traumas. The mechanisms of injury included falls from a height (n = 12; 11 suicidal and 1 accidental), traffic accidents (n = 3), and crush injuries (n = 1). The mean ISS was 25.94 ± 13.88 (range: 9-50). Thirteen of the 16 patients showed associated injuries, including 9 cases of associated injuries of the head and trunk with an Abbreviated Injury Scale Score ≥3. Other injuries of the spine and extremities were present in 10 patients, while two patients experienced pelvic fracture and urinary tract injuries. At the time of initial examination, neurological deficits were observed in 12 out of 16 patients, with a mean Gibbons' grade of 3.13 ± 1.25 (range: 1-4).
3.6. Complications. All patients showed bone union. No case of deep and/or superficial wound infection was noted. Screw loosening was observed in 3 patients; one patient had greater reduction loss (case 3; bilateral S1 and S2 IS screws, see Figure 5). Further, screw malposition was observed in 4 patients; however, in these cases, no new neurological deficits were observed postoperatively. Advances in Orthopedics 5

Discussion
The present study aimed to show that closed reduction in a hyperextended supine position with manipulations and percutaneous TSTI/IS screw fixation is useful for treating Denis zone III sacral fractures. Currently, the present study is the largest case series regarding closed reduction and percutaneous screw fixation for these fractures [8,10,11,13]. Furthermore, the present study includes the greatest number of cases involving the supine position maneuver. Denis zone III sacral fractures may be H-, U-, T-, or Y-shaped [20], and sacral dislocations may be anterior or posterior [1]. When treating these fractures, it is important to reduce the sacral kyphotic angle fracture translation and to maintain this reduced position. Previously, the outcomes of several treatments for these fractures have been reported. Siebler et al. reported the results of conservative treatment [21], and found that the sacral kyphotic angle increased by 4.1 ∘ (from 36.4 ∘ to 40.5 ∘ ) posttreatment. In terms of operative treatment, Schildhauer et al., Tan et al., and Lindahl et al. reported that using spinopelvic fixation, the initial kyphotic angles of 43 ∘ , 32 ∘ , and 38 ∘ were improved to the final angles of 21 ∘ , 19 ∘ , and 22 ∘ , respectively [6,20,22]. Nork et al. reported that, after using IS screw fixation, the kyphotic angle was reduced from 29.2 ∘ to 28.2 ∘ [8], whereas König et al. reported that, after using TSTI screw fixation in 3 patients, the sacrococcygeal and pelvic incidence angles were reduced from 84 ∘ and 75 ∘ preoperatively to 58 ∘ and 56 ∘ postoperatively, respectively (reduction losses of 14 ∘ and 15 ∘ , respectively) [10]. Ruatti et al. reported that, in a supine position, hyperlordosis, skeletal traction, and percutaneous IS screw fixation had good reduction results in 3 cases [13]. In the present study, the pre-and postoperative kyphotic angles were 40 ∘ and 22 ∘ , respectively, indicating a reduction loss of 4.9 ∘ . In terms of translation, Lindahl et al. reported that the mean pre-and postoperative translations with spinopelvic fixation were 15.5 and 5.8 mm, respectively [22], whereas in the present study, the corresponding values were 8.9 and 2.8 mm. In the present study, one patient with IS fixation only (Case 3) was observed to have a high reduction loss and screw loosening. Taken together, these clinical data suggest that TSTI screw fixation is equal to spinopelvic fixation. Furthermore, Min et al. reported that 2-TSTI screw fixation was superior to spinopelvic fixation in a zone 2 sacral fracture model from a biomechanical viewpoint [23]. Therefore, we believe that TSTI screw fixation in the supine position is an effective, rigid, and minimally invasive procedure for Denis zone III sacral fracture fixation.
Denis zone III sacral fractures have been reported to result in neurological injury of varying severity in up to 100% of patients [6-8, 20-22, 24]. Accordingly, in the present study, neurological deficits were seen in 12 of 16 patients (75%). However, neurological recovery has been shown to occur in 50-100% of patients with these fractures [6-8, 10, 20, 21]. Early surgical treatment of the sacrum, including restoration of spinopelvic stability and decompression of the nerve roots, Advances in Orthopedics 9 indirectly or directly, is thought to provide the best possible environment for neurological recovery. Schildhauer et al. reported that neurological recovery depends on the extent of nerve damage at the time of injury [6]. However, Siebler et al., in their study of nonoperative treatment, reported that the recovery rate was still as high as 85.7%. [21] Thus, the best treatment method, in terms of neurological outcomes, remains unclear. In the present study, the overall neurological recovery rate was 66.7% (8/12). A lack of neurological recovery was observed in four cases (cases 3, 4, 12, and 15). Of these, one case (case 4) involved a spinal cord injury merger. In an additional case (cases 12), decompression of the postoperative sacral canal was deemed sufficient; however, no neurological recovery was observed. In cases 3 and 15, because of a suspected reduction loss in the kyphotic angle and translation, additional nerve decompression surgery was performed (case 3); however, nerve recovery could not be obtained. In addition, four cases (cases 6, 7, 8, and 9; see Figure 4) underwent additional decompression because of remnant fragment and/or insufficient neurological recovery and were finally considered to have sufficient neurological recovery. Based on our experience and the report by Schildhauer et al. [6], the likelihood of neurological recovery appears to be dependent on the degree of nerve damage at the time of injury. However, for maximum neurological recovery, we consider indirect or direct nerve decompression to be necessary. Hence, in our treatment strategy, indirect reduction and decompression are performed in the acute phase and if the postoperative CT shows incomplete decompression and/or remnant fracture fragments in the sacral canal, secondary direct sacral decompression is consequently performed.
We consider TSTI/IS screw fixation to be a less invasive method, associated with a relatively low complication rate. In our series, the mean operative time was approximately 65 minutes. In cases involving TSTI or IS screw insertion only, the screw insertion time was approximately 10 minutes per screw. In addition, the time required for our preoperative closed reduction procedure is about 20 to 30 minutes. Hence, in cases requiring only closed reduction and screw fixation, patients can be treated in the acute phase. Furthermore, in cases involving screw fixation only, the patient is positioned supine, which is often the most comfortable position for polytrauma patients. In addition, TSTI/IS screw fixation is associated with a very low profile and few surgical site complications. In fact, in our study, no surgical site complications were observed. In past reports using different approaches, the surgical site complication rate was as high as 38% [6,7,20]. Spinopelvic fixation is associated with a particularly high complication rate in pelvic ring fracture treatment [6,20], and posterior soft tissue complications are associated with poorer outcomes.
There are several limitations to the present study. First, the study population was relatively small and did not allow for a highly powered statistical analysis; therefore, our conclusions should be interpreted with caution. Second, there are operative limitations for TSTI/IS screw fixation. In TSTI/IS screw fixation for zone III sacral fractures, at least two screw insertions (2 TSTI screws, or 1 TSTI screw and 1 IS screw) are required, as the screws need to provide resistance against the vertical load shear force and the rotational force of the sacral vertebral body. However, it has been reported that ilio-sacroiliacal corridors for intraosseous implants were not inserted in 18-25% and 10-12% of S1 and S2 cases, respectively; moreover, a high frequency of this issue has been reported for female patients [25][26][27]. Thus, owing to anatomical variance and different fracture types, this method is not always useful.
In conclusion, the current study showed that closed reduction in a hyperextended supine position with TSTI/IS screw fixation is effective for treating Denis zone III sacral fractures in terms of the fracture reduction, loss of reduction, neurological recovery, and clinical outcomes. Thus, despite some limitations, we believe that our procedure is an effective and appropriate method for treating Denis zone III sacral fractures.

Data Availability
The data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest
None of the authors has received any financial or personal support from other people or organizations that could inappropriately influence this work.