An early experience of lumbopelvic fixation techniques at Alexandria University

The aim of lumbopelvic fixation is to obtain a solid fusion across the lumbosacral junction. There are many indications for lumbopelvic fixation, namely, spinal deformity in cases requiring long segment fusion, pelvic obliquity, pseudarthrosis at the lumbosacral junction, infection or osteolytic tumors, and pathologic fractures. The classical iliac screws should be contained within the iliac bone but have some disadvantages: excessive soft tissue dissection needed for accurate insertion, screw prominence with patient discomfort, and usually, a side connector is needed to connect the iliac screws to the rest of the construct. Lumbopelvic fixation by insertion of S2 alar-iliac (S2AI) screws was recently described to overcome these disadvantages. In this study, the authors present the initial results for the evaluation of lumbopelvic fixation through the insertion of S2AI screws in 19 consecutive patients operated in the neurosurgery department at Alexandria University. The aim of the study was to evaluate the efficacy and complications of lumbopelvic fixation through the use of S2 alar-iliac screws. The authors conducted a retrospective cohort study of data collected from the database of patients who underwent lumbopelvic fixation through the insertion of S2AI screws from 2016 to 2019 at a single institution. There were 19 patients indicated for lumbopelvic fixation, operated by modern instrumentation systems using lumbar pedicle screws and S2 alar-iliac screws. There were 14 females and 5 males. The mean age at the time of the operation was 38.6 ± 19.4 years with a range from 11 to 65 years. There was a total of 37 S2AI screws, screw diameter was 7mm in all cases regardless of age, and the length of the screws ranged from 50 mm in a young female patient (11 years) to 90 mm in an old male patient (60 years). Two screws were inserted per patient except in one case with congenital scoliosis due to the bad bone quality and the multiple iatrogenic wrong paths. Postoperative VAS score for back pain was greatly improved in all patients after the first 6 months of follow-up from 8 ± 1.5 to 3.5 ± 1.2 (paired t-test = 11.182, P<0.001). All patients had a good spinal range of motion to maintain normal daily activities without any significant restrictions after the first 3 months of follow-up. Immediate postoperative radiological follow-up had revealed 2 cases of posterior pelvic breaches and one case with anterior pelvic breach but without clinical manifestations with no need for revision. Two cases of unilateral screw breakout were observed after the first 6 months of follow-up. Removal of screws after the first 6 months was done in one patient with spondylodiscitis due to the unresolved infection and screw pullout. The insertion of S2AI screws is an effective technique for lumbopelvic fixation with a relatively low rate of complications. Pelvic breaches are the commonest complications encountered during the insertion of S2AI screws, although no significant clinical morbidities were reported.


Introduction
The main aim of lumbopelvic fixation is to obtain a solid construct across the lumbosacral junction. The main indications of lumbopelvic fixation include the following: spinal deformity in cases requiring long segment fusion and in cases with high pelvic obliquity, pseudarthrosis, infections or tumors at the lumbosacral region associated with bone loss, and pathologic fractures [1,2]. Classically, the starting point for pelvic screw insertion lies close to the posterior inferior iliac spine and should be contained within the boundaries of the thick iliac bone and has been shown to obtain a stable lumbopelvic construct [3]. However, lumbopelvic fixation through the use of the classical iliac screws has some disadvantages: excessive soft tissue dissection is needed for accurate insertion, soft tissue irritation and patient discomfort associated with the prominence of the classical iliac screws, and finally, the usual need of side connectors to connect the iliac screws to the remainder of the construct [4,5]. In order to avoid most of these complications, a novel way for lumbopelvic fixation has evolved [6,7]. Chang et al. first described lumbopelvic fixation by insertion of S2 alar-iliac (S2AI) in 2009 [8]. Although the freehand techniques for the insertion of most spinal implants have been evaluated with encouraging results [9,10], however, a paucity of data exists with regard to the evaluation of the free-hand technique for S2 alar-iliac screw insertion [6]. In this study, the authors present the initial results for evaluation of lumbopelvic fixation through the insertion of S2AI screws in 19 consecutive patients operated in the neurosurgery department at Alexandria University.

Methods
The authors conducted a retrospective cohort study of data collected from the database of patients who underwent lumbopelvic fixation through the insertion of S2AI screws from 2016 to 2019 at a single institution.
The aim was to evaluate the efficacy and complications of lumbopelvic fixation through the use of the S2 alariliac screws. There were 19 patients indicated for lumbopelvic fixation, operated by modern instrumentation systems using lumbar pedicle screws and S2 alar-iliac screws. There were 14 females and 5 males. The mean age at the time of the operation was 38.6 ± 19.4 years with a range from 11 to 65 years. The indications for lumbopelvic fixation were the following: 7 cases of spondylolisthesis (3 cases of isthmic spondylolisthesis and 4 cases of degenerative L5-S1 spondylolisthesis), 6 cases of scoliosis (1 case of congenital lumbar hemivertebra, 2 case of degenerative scoliosis, and 3 cases of adolescent idiopathic scoliosis), 2 cases of fracture sacrum, 2 cases of L5-S1 discitis, one case of degenerative retrolisthesis, and one case of sacral tumor ( Table 1). The main indications for pelvic fixation in the 4 cases of degenerative spondylolisthesis were the poor bone quality (osteoporosis) in the affected patients with the need for fusion of multiple motion segments down to the sacrum. The three cases with isthmic spondylolisthesis were associated with high degree spondylolisthesis. The presence of a long construct extending down to the sacrum in scoliosis cases was the main indication to include the pelvis in the construct. The eroded S1 sacral body was the main indication of pelvic fixation in the 2 cases of L5-S1 discitis and the case diagnosed with the sacral tumor.
There was one case of spontaneous L5-S1 discitis in a drug addict adult, while the second case was an old diabetic male suffering from postoperative L5-S1 discitis following an operation for lumbosacral decompression and fusion from L3 down to the sacrum. The case of postoperative discitis was complaining of persistent severe low back pain for 3 years after the initial surgery. The radiological investigations had revealed the signs of L5-S1 discitis in association with a hardware failure in the form of sacral screws pulled out. The case with the degenerative retrolisthesis was complaining of severe low back pain and bilateral sciatica following an operation for lumbosacral decompression and fusion from D11-L4. The patient was complaining of back pain and sciatica for 1 year after the initial surgery. The radiological investigations had revealed L4-5 instability with retrolisthesis with the necessity to extend the fusion down to the sacrum and pelvis. All patients were operated by senior spine surgeons for lumbopelvic fixation via the insertion of S2-alar iliac screws which were connected with the remaining lumbar pedicle screws via the same rods ( Fig. 1). Postoperative radiological evaluation was done by PXR and CT scans. Evaluation of the patients was done via clinical and radiological follow-up for at least 9 months and up to 18 months.

Outcome measures
Spinal range of motion and VAS score for back pain were used for clinical evaluation of the involved patients. Radiological assessment was obtained via plain X-ray AP and lateral views of the instrumented segments in addition to the CT scanning for more detailed evaluation. Data were fed to the computer and analyzed using the IBM SPSS software package version 20.0. (Armonk, NY: IBM Corp). The Kolmogorov-Smirnov was used to verify the normality of distribution of variables. Paired t-test was assessed for comparison between two periods for normally distributed quantitative variables. Significance of the obtained results was judged at the 5% level.
An approval from the research ethics committee of the Faculty of Medicine, Alexandria University [serial number 0304468], was obtained in November 2019.

Surgical technique
After subperiosteal dissection of the paraspinal muscles attached to the lumbosacral spine, lumbosacral pedicle screws were inserted by the traditional fluoroscopyguided methods. The entry point for insertion of S2 alar-iliac screws was revealed by more distal dissection along the sacrum, and it should be in line with S1 pedicle screws or just lateral at any point between S1 and S2 dorsal foramina. It was easier to insert the S2 alar-iliac screws from the contralateral side by initial drilling of the entry point followed by tapping using a 2-mm blunttipped pedicle probe. The pedicle probe was advanced towards the anterior inferior iliac spine crossing the hard surface of the sacroiliac joint with the curve of the probe directed dorsally to avoid anterior perforation of the pelvic wall. Once the hard surface of the sacroiliac joint was reached at a depth of about 50 mm in adults, the pedicle probe was then removed, and assessment of the osseous channel by a blunt ball-tipped flexible sound was done. The assessment should include the four osseous walls beside the floor. The pedicle probe was reinserted once more with the curve of the probe directed ventrally and advanced towards the anterior inferior iliac spine in a smooth consistent manner. AP view by C-arm was a guide to avoid breaching into the sciatic notch or the acetabulum. Once the pedicle probe reached the thick iliac bone above the sciatic notch, it was removed, and the osseous channel was reassessed once again to make sure a secure path for the screw and to assess the t, paired t-test; p, p value for comparing between pre and post 6 months *Statistically significant at p ≤ 0.05 Fig. 2 a Preoperative LSS CT scan showing a case of congenital scoliosis in a 55-year-old female patient with a left-sided hemivertebra in between L4 and L5. b Immediate postoperative plain X-rays after removal of the hemivertebrae and partial correction with lumbopelvic fixation at one side desired screw length which usually ranges from 70 to 90 mm in adults. The screw was then inserted within the desired path, and its position could be confirmed by C arm. A rod was inserted to attach the S2AI screw to the remainder of the lumbosacral construct without the need of any side connector.

Results
There were 19 patients indicated for lumbopelvic fixation, operated by modern instrumentation systems using lumbar pedicle screws and S2 alar-iliac screws. There were 14 females and 5 males. The mean age at the time of the operation was 36.4 ± 19.4 years with a range from 11 to 65 years. There was a total of 37 S2AI screws, screw diameter was 7mm in all cases regardless the age, and the length of the screws ranged from 50 mm in a young female patient (11 years) to 90 mm in an old male patient (60 years) ( Table 1). All the S2AI screws were polyaxial. Two Screws were inserted per patient except in one case with congenital scoliosis due to the bad bone quality and the presence of multiple iatrogenic wrong paths (Fig. 2). Postoperative VAS score for back pain was greatly improved in all patients after the first 6 months of followup from 8 ± 1.5 to 3.5 ± 1.2 (paired t-test=11.182, P< 0.001) ( Table 2). All patients had a good spinal range of motion to maintain normal daily activities without any significant restrictions after the first 3 months of followup. No complications related to the screw prominence were recorded. Radiological follow-up had revealed 2 cases of posterior pelvic breaches and one anterior pelvic breach but without clinical manifestations with no need for revision (Figs. 3, 4). Two cases of unilateral screw breakout were observed after the first 6 months of follow-up (Fig. 5). Removal of the screws after the first 6 months was done in one patient with spondylodiscitis due to unresolved infection with pelvic screw pull out and rod loosening (Fig. 6).

Discussion
Surgical fusion across the lumbosacral region is still considered a surgical challenge despite the availability of modern techniques of spinal fixation. The sacrum has a complex anatomy and poor bone quality; in addition, there is a unique biomechanical force across the lumbosacral junction. These factors greatly contribute to the high failure and complication rates associated with fusion across the lumbosacral junction [11][12][13].  Augmentation of lumbosacral fusion through the use of pelvic screws is needed to decrease the high failure and complication rate associated with the lumbosacral fixation techniques [12,13]. Classically, the traditional iliac screws have been used effectively to augment lumbosacral fusion. However, there are some drawbacks that may limit the use of the traditional iliac screws: the excessive soft tissue dissection needed before the screw insertion, the need for a side connector to connect the iliac screws to the rest of the lumbosacral construct, and finally, the usual complain of postoperative patient discomfort associated with the screw prominence [4,5,13]. As a result of these disadvantages, many spine surgeons are discreet to use the classical iliac screws, and hence, the more recent use of the S2 alar-iliac screws [8]. Many authors have demonstrated better clinical and radiological results with the use of the S2AI screws. Sponseller et al. reported few complications with the use of the S2AI screw technique [14]. Ilyas et al. also reported better results with S2AI screws in adult and pediatric populations compared with the traditional iliac screw insertion [5]. The freehand technique for insertion of S2 alar iliac screws has recently been described and evaluated by few authors [6,15]. This technique has the advantage of easily accessible anatomical landmarks: the entry point that is easily inspected without excessive soft tissue dissection, and the target point, the anterior inferior iliac spine, which is easily palpable. Although does not completely obviate the use of intraoperative fluoroscopy, this technique could lead to less exposure to harmful radiation for both the surgeon and the patient, in addition to less operative time with better outcome [6,16]. Based on our results, we report that the use of S2 alar-iliac screws is an efficient method for lumbopelvic fixation. Our series revealed 3 major radiological complications, namely, pelvic breaches, screw breakout, and screw pull out (Figs. 3, 4, 5, and 6). There were 2 breaches in the posterior pelvic wall (10.5%) and one breach through the anterior pelvic wall (5.3%) but without any clinical morbidity. O'Brien et al. reported a breach rate of 15% through the posterior pelvic wall muscles in a cadaveric study [17]. Shillingford et al. found a rate of 7% breach through the posterior pelvic wall and only 1% breach through the anterior wall but without any major complications [6]. Inferior violation into the sciatic notch was not encountered in our series. This type of violation may result in serious complications related to the superior gluteal neurovascular bundle and the sciatic nerve as they pass through the sciatic notch [18]. Our results had revealed 2 cases of unilateral screw breakout (10.5%) within the first year of follow-up. This was observed in patients with high degree isthmic spondylolisthesis and degenerative scoliosis. None of these patients has been returned to the operative room for re-do surgery, as none of them experienced a significant change in back pain VAS score. The screw breakout complications may be due to the use of the polyaxial S2AI screws rather than the monoaxial screws. Another possible cause for the screw failure is the acute angle between the screw head and the screw shaft of the S2AI screws. Also, in the case operated for correction of highgrade isthmic spondylolisthesis, the high degree of the sacral slope may increase the biomechanical stresses across the region of the lumbosacral junction with more stresses on the pelvic screws (Fig. 5). Although there might be a relation between the postoperative sagittal balance and the S2AI screw failure, but to date, there is limited data evaluating the sacropelvic fixation using the S2AI screws in terms of postoperative sagittal balance outcomes [5,13,19]. In 2015, Guler et al. had reported a multicenter review of 45 patients who underwent sacroiliac fixation with S2AI screws (44%) or the classical iliac screws (56%). The authors reported a failure rate of 35% for the S2AI group, while the failure rate among the iliac screw group was only 12%. This difference was statistically significant. The authors acknowledged this high failure rate to the use of the polyaxial S2AI screws as there was no failure reported with the use of the monoaxial S2A screws. They also reported the acute angle between the screw head and the screw shaft as a potential cause of failure [13]. Screw pull out and rod loosening were observed in one patient (5.9%) with spondylodiscitis and unresolved infection. This particular patient was operated for removal of the whole construct due to pseudarthrosis after the first 6 months of follow-up (Table 3).

Study limitations
The small numbers of the cohort group with a relatively short period of follow-up are the main limitations of this study.

Conclusion
The insertion of S2AI screws is an effective technique for lumbopelvic fixation with a relatively low rate of complications. Pelvic breaches are the commonest Fig. 6 a Three years postoperative lateral view plain X-ray image of a 65-year-old diabetic male patient operated for lumbosacral fixation, showing the signs of L5-S1 discitis and hardware failure in the form of sacral screw pull out. b Six months postoperative plain X-ray AP and lateral views of the lumbosacral spine region after removal of the old hardware and insertion of a new hardware from L3 down to the pelvis escaping the L5 and S1 bodies due to the bad bone quality and the excessive osteolytic reaction associated with the infective process. Note the pull out of the pelvic screw and rod loosening complications encountered during the insertion of S2AI screws, although no significant clinical morbidities were reported.