With the tremendous advances in spinal instrumentation technique over the last quartercentury, reduction of HGS can now be accomplished more safely and effectively than ever before [2]. The primary rationale of reduction severe slip is to improve the sagittal malalignment through correcting the LSA and the patient’s ability to stand upright, with a secondary advantage of a reduction procedure being an improvement in fusion rate [20, 21, 22]. Partial reduction of slip is proved to be a safe and effective choice. First, HGS in adluts have often reached a stable position, auto-fusion or ankylosis of the slipped level can occur, resulting in difficulties in anatomy reduction. Second, most of the total L5 nerve strain occurs during the second half of reduction [23]. Third, partial reduction can also lead to correction of slip angle which is related to the risk of progression and is the key to restoring sagittal alignment [10, 24]. Moreau et al. [25] treated 50 patients with HGS through a single posterior approach with IS fixation and partial reduction. The mean listhesis grade reduced by >50%, the LSA was significantly improved. Seventeen patients (34%) showed postoperative radicular deficit. Hart et al. [26] treated 16 patients with modified Bohlman technique, slip percent reduced by 25%, one patient suffered L-5 radiculopathy. In these series of patients, slip percentage reduced by 31.2% (from average 62.71% to 31.54%). The SA and LSA significantly improved after surgery and remained stable at the last follow. One patient (4%) manifested right L5 nerve root palsy after surgery. In the premise of satisfied SA and SLA correction, proper less reduction of slip might have advantages of lower radicular injury rate.
Since Allen and Ferguson [27] introduced the landmark Galveston technique, the iliac screw fixation was then developed and was demonstrated to have superior biomechanical stability as compared to any of the preceding spinopelvic fixations [28]. However, wound complications, implant prominence and subsequent buttock pain are well-acknowledged drawbacks of IS technique [29, 30].
S2AI with deeper and medial insertion position, lower profile screw head and easier linkage with cephalad construct, was developed to be an alternative technique of traditional IS [30, 31]. S2AI is designed to purchase three-layer cortex. However, biomechanical study have shown that IS and S2AI constructs have demonstrated similar stability in terms of stiffness and load to failure [32, 33].
It is intriguing that the mode of implant failure may be different in IS and S2AI techniques. Shabtai et al [34] demonstrated that a higher implant failure rate in the traditional IS group was a result from the disengagement of rod-iliac screw connector. The iliac screw and the side connector attachment creates a large moment arm with the proximal construct and a potential weak point in the construct. The implant failure etiology for S2AI was different from the traditional IS. Guler et al. [35] reported that S2AI harboured a higher risk of short-term acute failure in comparison of IS (35% in S2AI screws vs 12% in traditional IS). They postulated that the acute angle developed between the screw head and shaft brought about a stress riser in the S2AI screw and caused it to fail when it crossed the sacral-iliac joint. According to Keorochana et al [16], traditional IS screw fixation had higher postoperative complications and revisions than did S2AI fixation in adult and pediatric populations. Ishida et al. [36] demonstrated that the IS technique have higher rate of overall reoperation than the S2AI technique, but interestingly, the majority of reoperations in the IS group were attributable to surgical site infection. Similarly, Elder et al. [37] also concluded that the higher rates of revision surgery was caused by iliac screw head being prominent in the subcutaneous tissue, causing symptoms such as hip or buttock pain, may in turn lead to wound complications breakdown and infection.
We modify the traditional IS technique in a simplified fashion to deal with the drawbacks mentioned above. First, we use the lumbar pedical screw with maximum size (6.5mm in diameter, 55mm in length), which is smaller than the traditional IS (8mm in diameter and 80-100mm in length), thus we can choose a more inferior and medial insertion point to further avoid hardware prominence. Additionally, due to the screw head is smaller than that of traditional IS, the embeddedness of screw head is deeper and in turn increase effective length of 1cm. Second, we use the traditional rod instead of off-set connector with easy rod contouring technique through commonly used bender to simplify the procedure. The rod contour between S1 and IS screw is smooth and obtuse-angulate, which is similar to the physical curvature, maintaining the original rod rigidity. Moreover, the length of rod between S1 pedical screw and IS is short, which shortening the moment arm and providing constructive stability to resist torsion and shear movement. In the present study, one patient (4.1%) suffered wound infection, which is lower than the previous reports of 4.2-20% [38, 39, 40]. The VAS and ODI scores were significantly improved at the last follow-up, no patient suffered severe hip or buttock pain.
We performed unilateral IS fixation in all patients. There are several studies that comparing the outcomes between bilateral and unilateral IS fixation. Saigal et al. [41] demonstrated there was no added benefit with bilateral iliac screw fixation over unilateral iliac fixation in most cases and it would led to comparable rates of complications. Tomlinson et al. [42] reported that there were no biomechanical differences between bilateral and unilateral iliac screw fixation, and clinical evidence has reported better outcomes and fewer complications when using unilateral fixation. In these series of patients, the fusion rate was 95.8%. One patient with rheumatoid arthritis and severe osteoporosis was diagnosed to be fusion failure and underwent the revision surgery. Three patients (12.5%) manifested asymptomatic IS loosening after fusion has occurred. The outcomes proved that the modified unilateral IS fixation was able to provide enough fixation stability.
Historically, PI was considered a constant anatomical parameter after maturity in the absence of pelvic fractures or sacropelvic tumor resection [43, 44]. Actually, the motion of sacroiliac joint (SIJ) which was affected by sitting, supine position or others locomotion activities, would cause the change of PI. The magnitude of SIJ movement in adulthood has been reported ranging from 1 to 4° of rotation [45]. Additionally, the laxity of SIJ due to degeneration is considered to be the fundamental of PI change [46]. According to Dreyfuss et al. [47], SIJ has less resistance to rotation force during the motion. Thus, we postulate that: on the one hand, the patient positioning of slight hip extension during surgery might cause the anterior rotation of the sacrum, resulting in intraoperative increase of PI, which would be finally maintained by the application of sacroiliac fixation, on the other hand, incremental stress to SIJ after spinopelvic fixation accelerates SIJ degeneration, inducing an increasing motion. In this study, we find the PI is increased by an average of 4.3° at the final follow-up, which is smaller than the findings of Lee et al. (5.9°) [48] and larger than the results of Ishida et al. (2.9°) [36]. The dramatically decrease of PI after S2AI fixation was well-introduced by many authors [19, 36, 48, 49]. The change in PI after S2AI instrumentation may be due to direct fixation and modification of the SIJ, which have been reported by cadaveric study [50]. From our point of view, the IS would have advantage over S2AI regarding the PI change. An obvious decreased PI value may potentially affect the self-adjust ability of sagittal balance at spinopelvic area, as the SIJ is the last mobilizable joint after lumbosacral fusion. Moreover, a pelvic with larger PI would have greater compensation ability than the pelvic with small PI [51]. However, postoperative changes in spinopelvic parameters are a complicated phenomenon. The mechanism of how spinal alignment may be affected by instrumentation and how it is translated into clinical outcomes is still unknown, further study is needed.
There are several limitations in this study. First, the sample size was small, involving only 32 patients. Larger sample sizes will be needed to provide stronger evidence to our conclusions. Second, the follow-up period was short with an average of 49 months. Prolonged follow-up was needed to figure out the mechanism of PI change and the consequent long-term outcomes. Finally, we performed unilateral IS fixation in all patients. The clinical and radiological outcomes by now proved that the constructs was able to provide enough fixation strength for reduction and stability, however biomechanical study would be of great necessity.