Learns more about the ioverame dials branch treatments

S2-Alar-Iliac Pelvic Fixation With Standard vs Porous Fusion/Fixation Implants: An Inverse Probability Weighted 1-Year Preliminary Comparative Analysis

  • International Journal of Spine Surgery
  • September 2026,
  • 8957;
  • DOI: https://doi.org/10.14444/8957

Abstract

Background Recently, porous fusion/fixation implants (PFFIs) with a 3D-printed porous, fenestrated, bone self-harvesting design were developed for S2-alar-iliac (S2AI) spinopelvic fixation. Prior studies have demonstrated the safety and feasibility of using PFFI for S2AI spinopelvic fixation, but their impact on outcomes remains unknown. This study compared 1-year outcomes of PFFI vs standard threaded screws.

Methods This is a single-center retrospective cohort study evaluating consecutive patients who underwent S2AI spinopelvic fixation between 2019 and 2024. Patients who underwent S2AI fixation for long-segment fusion, high-grade L5–S1 spondylolisthesis, L5–S1 pseudarthrosis, and lumbosacral junction fractures were included. Patients with iliac fixation, prior spinopelvic fixation, injection-proven sacroiliac joint dysfunction, or lack of 1-year clinical and radiographic follow-up were excluded. Baseline demographic, clinical, and radiographic variables were collected. The primary outcome was distal failure. Patients were stratified by S2AI implant design. Inverse probability weighting was then performed to adjust for confounding and cohort imbalance, followed by comparative analysis of 1-year outcomes.

Results Ninety patients were included. Median age (interquartile range [IQR]) was 67 (61–74) years, including 59 (65.6%) women. Median (IQR) number of levels fused was 6.5 (4.0–9.0). There were 33 (36.7%) and 57 (63.3%) patients with PFFI vs standard S2AI implants, respectively. Overall distal failure and re-operation rates were 12.2% and 5.6%, respectively. Injection-proven sacroiliac joint dysfunction following S2AI fixation occurred in 3.3%. PFFI implants were significantly larger in diameter and length, but there were no other significant differences in baseline demographics, spinopelvic parameters, fusion levels, osteotomies, L5–S1 interbody technique, and bone morphogenetic protein-2 use between cohorts. After IPW, there were no significant differences in distal mechanical implant failure, L5–S1 pseudarthrosis, and sacroiliac joint dysfunction at 1 year. The PFFI cohort had a significantly lower rate of pelvic screw loosening (P = 0.011).

Conclusions In this preliminary short-term (1 year) analysis, PFFI and standard threaded implants demonstrated comparable rates of distal failure and sacroiliac joint dysfunction. Studies evaluating longer-term (≥2 years) clinical and radiographic outcomes are required.

Clinical Relevance PFFIs used for S2AI spinopelvic fixation demonstrated similar short-term 1-year outcomes compared with standard threaded implants, with a lower rate of pelvic screw loosening.

Level of Evidence 3.

Introduction

Spinopelvic fixation has been employed for several decades. Early biomechanical studies supported its role in promoting arthrodesis at the lumbosacral junction by stress-shielding S1 screws.1,2 More recently, S2-alar-iliac (S2AI) fixation has shown similar biomechanical properties as iliac fixation3,4 with an overall lower reported complication profile5,6 and lower overall reoperation rates compared with the iliac technique.6,7 A recent study found no differences in distal failure between S2AI and iliac techniques,8 although the authors found that the iliac cohort had a higher rate of revision surgery for distal failure indication. Sacroiliac joint dysfunction has also become an increasingly important consideration following spinopelvic fixation, with S2AI and iliac techniques demonstrating similar rates of posttreatment sacroiliac dysfunction.8 Pelvic screw loosening has been included in a recent classification scheme of distal failure, which we sought to incorporate into the present study.9 Alongside increased utilization of S2AI spinopelvic fixation technique, newer implant technologies have emerged.

Recently, a porous fusion/fixation implant (PFFI) with a 3D-printed bone self-harvesting design was developed (iFuse Bedrock Granite, SI-Bone, CA, USA) and approved by the US Food and Drug Administration. Its fenestrated structure is designed to facilitate biological osseointegration, which could confer potential advantages beyond conventional spinopelvic fixation techniques by further augmenting bony fusion across L5–S1 and possibly the sacroiliac joint itself. Initial clinical experience has demonstrated technical safety and feasibility of PFFI placement in spinopelvic constructs.10,11 Furthermore, short-term analysis (mean time to outcome of 7.1 months) of the manufacturer’s postmarket surveillance database found PFFI had comparable complication rates with standard threaded implants.12

In the present study, we performed an inverse probability weighted, head-to-head preliminary comparison of short-term (1 year) surgical outcomes in patients who underwent spinopelvic fixation with PFFI vs standard threaded implants.

Methods

This was a retrospective analysis from a single tertiary care center. This study was approved by the Institutional Review Board with consent exemption under 45 CFR 46.101(b).

Patient Selection

A query using the current procedural terminology code for pelvic fixation cases between 2019 and 2024 was performed to identify potential patients. Consecutive patients who underwent index spinopelvic fixation with the S2AI technique for the following indications were included: long-segment fusion (>3 fusion levels), high-grade L5–S1 spondylolisthesis (grade II or higher), prior L5–S1 pseudarthrosis, and lumbosacral junction nonpathological fractures (ie, not related to tumor or infection). Patients with less than 1 year of clinical and radiographic follow-up, prior history of spinopelvic fixation, more than 1 pelvic screw per side during index surgery, prior or concomitant sacroiliac joint fusion at the time of index surgery, preoperative sacroiliac joint dysfunction (confirmed by a positive response to sacroiliac joint injection), inadequate pre- and postoperative (6 weeks following surgery) standing radiographs, and surgical indications for neoplasm or infection were excluded. Our inclusion/exclusion criteria are detailed in Figure 1.

Flowchart illustrating inclusion and exclusion criteria.
Figure 1

Flowchart illustrating inclusion and exclusion criteria.

Patient demographics and medical comorbidities, including age, sex, body mass index (BMI), smoking status, opioid use, diabetes, and radiographic bone density (measured in Hounsfield units), at the mid-point of the S1 vertebral body were recorded.13 Surgical variables, including surgical indication, number of fusion levels, prior and concurrent interbody technique at L5–S1 at the time of index S2AI fixation, technique and number of osteotomy levels, utilization of multiple rods across the lumbosacral (L5–S1) or lumbopelvic junction (S1-pelvis), use of bone morphogenetic protein-2 at L5–S1, pelvic screw diameter/length, and S2AI implant design (PFFI vs standard threaded), were recorded.

The primary outcome was distal failure identified on computed tomography (CT) and/or x-ray imaging at 1 year (imaging modality was recorded), including mechanical implant complications between L5–pelvis (eg, rod fracture below L5, S1 and/or pelvic screw fracture, set screw dissociation), L5–S1 pseudarthrosis in the absence of implant fracture or disconnection (eg, S1 screw loosening, cage subsidence, absence of L5–S1 facet ankylosis, or bone growth across the L5–S1 interspace), and pelvic screw loosening.9 Pelvic screw loosening was assessed using the Shimizu grading scale.14 Patients with high-grade (Shimizu 2 or 3) loosening upon review of imaging by 2 independent reviewers were counted as pelvic screw loosening. Other outcomes included sacral insufficiency fractures, wound dehiscence or infection, reoperation for any indication, and subsequent sacroiliac joint dysfunction.

Assessment of Sacroiliac Joint Dysfunction

Patients presenting with lower back pain after index S2AI pelvic fixation underwent screening for sacroiliac joint dysfunction by the Fortin Finger Test and were assessed for tenderness to palpation over the sacral sulcus near the posterior superior iliac spine.15,16 If positive for pain in the sacroiliac joint region, the standard 5 provocative maneuvers for sacroiliac joint dysfunction were performed (Faber, compression, thigh thrust, distraction, and Gaenslen).17,18 If 3/5 tests were positive, patients were referred for sacroiliac joint injection. If injection provided a clinically significant reduction in pain as determined by pre- and postinjection scoring by the pain medicine team (>75%),19 a discussion was held with the patient regarding further management, including physical therapy, repeat injections, radiofrequency ablation, or sacroiliac joint fusion.

Spinopelvic Parameters

Spinopelvic parameters were measured on standing pre- and postoperative (6 weeks following surgery) outpatient radiographs by a senior author who was blinded to patient outcomes. Pelvic incidence (PI), pelvic tilt, lumbar lordosis (LL), and sacral slope were measured, and PI-LL difference was calculated (PI minus LL).

Baseline Statistical Analysis

Demographic, clinical, surgical, and radiographic categorical variables were reported using counts and percentages. Mean ± SD values were reported for normally distributed variables, and median (interquartile range [IQR]) values for skewed variables. Baseline variables and 1-year outcomes prior to propensity score weighting were compared for patients with either PFFI or standard S2AI implants using 2-way Fisher exact test or Pearson χ 2 test (categorical variables) and Student t test or Wilcoxon rank-sum test (continuous variables). P value <0.05 (2-sided for Fisher exact) was considered statistically significant. Additional univariate analyses were also performed for patients with available 2-year clinical and radiographic follow-up data as an exploratory and unadjusted analysis. All analyses were conducted using R software (version 4.5.0). As marginal associations may be confounded by pretreatment factors, inverse probability weighting (IPW) was applied to facilitate adjusted interpretation of the estimated effects.

Inverse Probability Weighting

To fully utilize available data and maximize statistical power, we employed IPW to compare the two S2AI designs and infer average treatment effects.20,21 Baseline demographic variables and potential confounders relevant to outcomes after spinopelvic fixation were included as covariates for IPW, including age, sex,22 BMI,23 number of fusion levels,8,24,25 and preoperative PI.26,27

The probability densities of propensity scores were plotted to evaluate overlap between cohorts. Standardized mean differences (SMD) were calculated to assess covariate balance before and after weighting, with SMD < 0.1 indicating an appropriate propensity score model and well-balanced covariates.21 After applying IPW, estimation of 1-year outcome differences was calculated. A 2-sided P value <0.05 was considered statistically significant. .

Sensitivity analyses were performed to identify whether the model was reproducible using different covariates. The first analysis used 2 additional covariates related to mechanical/surgical characteristics: new L5–S1 interbody and multi-rod construct spanning L5–S1 or S1–pelvis. The second analysis used 2 additional covariates related to baseline patient characteristics: smoking history and diabetes.

Results

Patient Characteristics

Two hundred sixty-three consecutive patients who underwent S2AI fixation during the study period were identified, and 90 patients met both inclusion and exclusion criteria (Table 1). Median (IQR) age was 67 (61–74) years (minimum 45 years, maximum 81 years). Sex was female in 59 (65.6%) patients, and mean BMI ± SD was 29.73 ± 6.02. There were 60 (66.7%) patients who never smoked, with the remaining being former or current smokers. There were 38 (42.2%) active opioid users. Twenty-four (26.7%) patients had diabetes. Median (IQR) Hounsfield units of the mid-S1 vertebral body were 171.50 (132–235) (n = 78). There were 13 (14.4%) and 2 (2.2%) patients who had prior L5–S1 interbody fusion with transforaminal lumbar interbody fusion or anterior lumbar interbody fusion, respectively. Eleven (12.2%) patients had pre-existing bony fusion at L5–S1 present on preoperative CT. Of those with prior interbody placement at L5–S1, 6 presented with nonunion at this level.

View this table:
Table 1

Demographics and clinical characteristics.

There were 69 (76.7%), 6 (5.7%), 9 (10.0%), and 6 (5.7%) patients who underwent S2AI fixation for long-segment fusion, high-grade L5–S1 spondylolisthesis, L5–S1 pseudarthrosis, and lumbosacral fractures, respectively. The median (IQR) number of fusion levels was 6.5 (4.0–9.0). Sixty-six patients (73.3%) had at least 1 level of posterior column osteotomy, and 57 (63.3%) had multilevel posterior column osteotomy during index S2AI fixation. Six patients (6.7%) had 3-column osteotomy, with 1, 4, and 1 patients with pedicle subtraction osteotomy at L2, L4, and L5, respectively. There were 41 (45.6%) and 12 (13.3%) patients with new L5–S1 interbody technique using transforaminal lumbar interbody fusion and anterior lumbar interbody fusion, respectively. Forty-one patients (45.6%) had bone morphogenetic protein-2 placed at L5–S1. There were 8 patients (8.9%) patients with multirod constructs spanning the distal junction (4 spanning L5–S1; 4 spanning S1–pelvis).

All patients underwent bilateral spinopelvic fixation with 1 S2AI screw per side. There were 33 (36.7%) and 57 (63.3%) patients with PFFI and standard threaded implants, respectively. In the standard cohort, 40 Reline (NuVasive, CA, USA), 7 Solera (Medtronic, CA, USA), 6 K2M (Stryker, MI, USA), 2 CREO (Globus Medical, PA, USA), and 2 Expedium (Depuy Synthes, Zuchwil, Switzerland) systems were placed. Pre- and postoperative spinopelvic parameters are presented in Table 1. At 1 year, 73 (81.1%) and 24 (26.7%) patients had available XR and CT imaging, respectively.

Clinical and Radiographic Outcomes

One-year outcomes are shown in Table 1. Distal failure was observed in 11 patients (12.2%). Of these, 4 patients (4.4%) had mechanical implant failures. The PFFI cohort had 1 pelvic set screw dissociation, 1 unilateral rod fracture at L5–S1. The standard cohort had 1 pelvic set screw dissociation and 1 pelvic screw shaft fracture. Four patients (4.4%) had L5–S1 pseudarthrosis without mechanical implant failures (PFFI cohort: 1; standard cohort: 3). Six patients (6.7%) had pelvic screw loosening (all in the standard cohort; 3 had isolated, asymptomatic pelvic screw loosening in the absence of L5–S1 pseudarthrosis and mechanical implant failure [fracture/dissociation]). All loose pelvic screws were Shimizu grade 2.14 One patient (1.1%) developed a sacral insufficiency fracture in the standard cohort, and 4 (4.4%) experienced wound dehiscence or infection. Five patients (5.6%) underwent reoperation within 1 year. In the PFFI cohort, reoperation was performed in 1 for mechanical implant failure and 1 for wound dehiscence/infection. In the standard cohort, reoperation was performed in 1 for L5–S1 pseudarthrosis in the absence of mechanical implant complications and 2 for wound dehiscence/infection.

Sacroiliac joint region pain was present in 10 patients (11.1%; positive tenderness to palpation and Fortin finger test). Provocative testing was positive in 4 patients (4.4%). Three patients (3.3%) were referred for and had a confirmatory positive response to sacroiliac joint injections, confirming the diagnosis of sacroiliac joint dysfunction. All 3 patients were managed conservatively without subsequent sacroiliac joint fusion surgery at 1 year follow-up.

Comparison of S2AI Implant Designs

Baseline characteristics for patients with PFFI and standard threaded implants were compared (Table 2). The PFFI cohort had significantly larger screw diameter (right and left, P < 0.001) and length (right and left, P < 0.001). There were no other statistically significant differences in baseline characteristics (including pre-/postoperative spinopelvic parameters and 1-year imaging modality).

View this table:
Table 2

Comparative baseline characteristics between PFFI and standard threaded S2AI implants.

Comparison of 1-year outcomes prior to IPW (Table 3) demonstrated no statistically significant differences for distal failure, sacral insufficiency fracture, wound dehiscence/infection, reoperation, or sacroiliac joint dysfunction. Pelvic screw loosening was higher in the standard cohort (10.5% vs 0%) but was not statistically significant (P = 0.082).

View this table:
Table 3

Comparative 1-year outcomes prior to inverse probability weighting between PFFI and standard threaded S2AI implants.

Inverse Probability Weighting

SMDs were calculated for selected covariates prior to and after IPW for the 2 cohorts (Figure 2). If unweighted, the SMDs for most covariates were large, indicating imbalance between the groups. After IPW, SMDs were substantially reduced, indicating well-balanced cohorts after propensity score adjustment. The propensity score densities (Figure 3) showed good overlap between treatment groups, supporting the validity of the IPW approach.

Standardized mean differences for covariates before (blue triangle) and after (red circle) inverse probability of weighting. BMI, body mass index; IPW, inverse probability weighting; PI, pelvic incidence; Sex1, female.
Figure 2

Standardized mean differences for covariates before (blue triangle) and after (red circle) inverse probability of weighting. BMI, body mass index; IPW, inverse probability weighting; PI, pelvic incidence; Sex1, female.

Propensity score densities comparing PFFI (red) and standard threaded (blue) implants after inverse probability weighting demonstrating good overlap. PFFI, porous fusion/fixation implant.
Figure 3

Propensity score densities comparing PFFI (red) and standard threaded (blue) implants after inverse probability weighting demonstrating good overlap. PFFI, porous fusion/fixation implant.

Comparative Analysis of 1-Year Outcomes

One-year outcomes after applying IPW are shown in Table 4. There were no significant differences for overall distal failure, distal mechanical implant failure, and L5–S1 pseudarthrosis without mechanical implant failure. However, the PFFI cohort demonstrated significantly lower frequency of pelvic screw loosening (subcategory of distal failure; P = 0.011). There were no significant differences for sacral insufficiency fracture, wound dehiscence/infection, reoperation (any indication), and injection-proven sacroiliac joint dysfunction. Sensitivity analyses incorporating additional covariates (Supplemental Tables 1 and 2) produced similar results.

View this table:
Table 4

IPW-adjusted 1-year outcomes stratified by S2AI implant type.

Standing AP radiographs of spinopelvic fixation with PFFI and standard threaded implants are displayed in Figure 4. Figures 4A–Bs demonstrate a patient with 10.5 mm diameter by 90 mm length PFFI implants at 6 weeks (A) and 1 year (B), with no radiographic evidence of distal failure. Figures 4C–D demonstrates a patient with 8.5 mm diameter by 80 mm length standard threaded implants at 6 weeks (C) and 1 year (D), with evidence of bilateral Shimizu grade 2 pelvic screw loosening. This patient was asymptomatic from his pelvic screw loosening.

Postoperative radiographs in a patient undergoing spinopelvic fixation with porous fusion/fixation implant (10.5 mm diameter by 90 mm length) at 6 weeks (A) and 1 year (B). Postoperative radiographs in a patient undergoing spinopelvic fixation with a standard threaded implant (8.5 mm diameter by 80 mm length) at 6 weeks (C) and 1 year (D) with evidence of bilateral pelvic screw loosening (Shimizu grade 2).
Figure 4

Postoperative radiographs in a patient undergoing spinopelvic fixation with porous fusion/fixation implant (10.5 mm diameter by 90 mm length) at 6 weeks (A) and 1 year (B). Postoperative radiographs in a patient undergoing spinopelvic fixation with a standard threaded implant (8.5 mm diameter by 80 mm length) at 6 weeks (C) and 1 year (D) with evidence of bilateral pelvic screw loosening (Shimizu grade 2).

Comparative Analysis of 2-Year Outcomes

Fifty-one patients had available 2-year clinical and radiographic data, with 14 (27.5%) in the PFFI cohort and 37 (72.5%) in the standard threaded cohort. Baseline univariate analysis (Supplemental Table 3) demonstrated the PFFI cohort had significantly larger implant diameter (right and left, P < 0.001). There were no other statistically significant baseline differences (including pre-/postoperative spinopelvic parameters and 1-year imaging modality). Head-to-head comparison at 2 years (Supplemental Table 4) demonstrated no statistically significant differences for distal failure, sacral insufficiency fracture, wound dehiscence/infection, reoperation, or sacroiliac joint dysfunction. Pelvic screw loosening was higher in the standard cohort (18.9% vs 0%) but was not statistically significant (P = 0.169).

Discussion

S2AI fixation technique was developed as a low-profile, in-line alternative to traditional iliac fixation requiring less soft tissue dissection. Here, we present an early experience comparing short-term (1 year) outcomes following S2AI pelvic fixation using PFFIs vs standard threaded screws. At 1 year postoperatively, overall complication and reoperation rates were low, with no significant differences in overall distal failure and injection-proven sacroiliac joint dysfunction. There was no difference in overall distal failure after IPW, including distal mechanical implant failure (fracture and/or dissociation) and L5–S1 pseudarthrosis with implant fracture/dissociation. However, after IPW, there was a significantly lower rate of pelvic screw loosening in the PFFI cohort compared with the standard implant cohort at 1 year. Notably, 50% of pelvic screw loosening cases had additional subcategories of distal failure, and 50% had isolated asymptomatic pelvic screw loosening without additional distal failure types.

Distal failure remains a major concern in spinopelvic fixation. Several systematic reviews and meta-analyses evaluating standard threaded S2AI implants reported lower implant complications and reoperation rates compared with iliac fixation.5–7,28 A multicenter retrospective review reported a 3.1% to 3.3% rate of acute S2AI fixation failure using standard implants (≤6 months of surgery), including set screw dissociation, screw fracture, rod fracture at the lumbopelvic junction, pelvic screw loosening, and sacral/iliac fractures.26 A single-center study of 291 patients who underwent standard threaded S2AI fixation reported only 1 patient (0.3%) with distal implant failure (pelvic screw breakage at 68 days).29 Recently, another study with ≥2 years of follow-up reported a 10.7% rate of distal failure.8 Here, we compared outcomes for distal failure between PFFI and standard S2AI implants. Notably, our study included patients with 1 pelvic screw placed per side, but multiple points of pelvic fixation have demonstrated lower rates of failure.29,30

Both standard threaded and PFFI S2AI techniques appear to fulfill the original goals of pelvic fixation2 by increasing stability at L5–S1, reducing stress on S1 screws.31,32 Here, there was an overall 12.2% rate of distal failure within 1 year. Of these, 4 patients (4.4%) experienced distal mechanical implant failure, including 2 (6.1%) in the PFFI cohort and 2 (3.5%) in the standard threaded cohort. One patient in the PFFI cohort underwent reoperation for distal mechanical implant failure, and the other 3 were managed conservatively at 1 year. In addition, 4 patients (4.4%) experienced L5–S1 pseudarthrosis without implant failure, with 1 patient (3.0%) in the PFFI cohort and 3 (5.3%) in the standard cohort. One patient in the standard cohort underwent reoperation for L5–S1 pseudarthrosis without implant failure, and the other 3 were managed conservatively at 1 year. In our study, IPW was applied to address imbalances in cohort sizes to estimate the effect of PFFI fixation. After IPW, there were no significant differences between cohorts for distal mechanical implant failure or L5–S1 pseudarthrosis without implant fracture/dissociation.

The one significant difference of this study was a lower frequency of pelvic screw loosening with PFFI fixation. For standard threaded S2AI fixation, reported rates of pelvic implant loosening are highly variable (0%–21.4%).24,33,34 Novel PFFI was designed to permit biological integration with host bone, potentially solidifying the bone-implant interface.12,35 Here, all 6 cases of pelvic screw loosening occurred in the standard threaded cohort considered Shimizu grade 2 loosening,14 with 50% of these cases being asymptomatic isolated findings in the absence of other distal failure categories (mechanical implant complications, L5−S1 pseudarthrosis without implant fracture/dissociation, or sacroiliac joint dysfunction). While pelvic screw loosening is included in recent classification schemes of distal failure,9 the clinical significance of isolated pelvic screw loosening remains incompletely understood and warrants further investigation. A recent study investigating this question found that delayed pelvic screw loosening was associated with significantly higher odds of revision pelvic fixation.36 The authors found about half of pelvic screw loosening cases did not require revision surgery. Our preliminary 1-year analysis also raises the question of whether/when isolated pelvic screw loosening is clinically significant, and longer-term follow-up is needed to determine whether a difference in revision surgery rates is appreciated comparing PFFI and standard threaded implants.

Sacroiliac joint dysfunction is an increasingly recognized problem after lumbosacral and lumbopelvic fusion.37 Here, we found no significant differences in postoperative sacroiliac joint region pain, positive provocative exam testing, and referral/positive response to sacroiliac joint injection following S2AI fixation with either implant design. S2AI fixation has been proposed to reduce sacroiliac joint pain,32 with 2 points of fixation demonstrating decreased motion compared with 1.31 Although S2AI fixation has been shown to reduce sacroiliac joint motion up to 50%, reported rates of postoperative sacroiliac joint dysfunction are highly varied, with reports ranging from 3% to 90%.8,38,39 In finite element analyses, peak stress on S2AI implants occurs at the sacroiliac joint in response to loading forces.4,40 In addition, the potential impact of S2AI fixation on subsequent bony fusion of the sacroiliac joint remains understudied.32 Biomechanically, novel PFFI fixation has demonstrated similar sacroiliac joint stability compared with standard threaded screws in finite element analysis investigation.41 Biologically, PFFI showed increased bone infiltration (osseointegration) in an ovine model,35 which could provide a potential benefit with respect to arthrodesis across the sacroiliac joint. Studies with longer follow-up, larger sample sizes, and dedicated imaging are needed to better characterize the role of novel implants for minimizing sacroiliac joint dysfunction and promoting de facto sacroiliac joint fusion.

Limitations

This study has several limitations. The methodology was retrospective, predisposing to bias and missing information. Despite IPW, sample sizes were relatively small, with the potential for unstable estimates. In addition, IPW adjustment was limited to a relatively small number of covariates given the sample sizes, and despite the use of clinically relevant covariates, there may remain unmeasured confounders within the data. In an attempt to address this problem, additional sensitivity analyses were performed with similar results. Follow-up time was limited to 1 year to assess short-term outcomes and may not accurately reflect/capture outcomes after pelvic fixation seen at later time points, and our 2-year assessment is limited in sample size. CT data were not available for the majority of patients at 1 year as surgeon preference varied, and we recognize plain radiographs used to determine distal failure may underestimate or misclassify outcomes. Longer-term follow-up with consistent CT imaging would provide more information on whether there is a difference in outcomes between PFFI and standard threaded implants. Apart from sacroiliac joint pain and provocative testing, patient-reported outcomes data were lacking. Furthermore, PFFI is a statistically larger implant (diameter and length) with a porous, fenestrated design that is mechanistically different compared with standard threaded screws. Due to the imbalance in implant size, sensitivity analysis including implant diameter or length was not feasible due to a binary estimate of propensity scores. Whether differences in S2AI implant size (separate from surface design) impart a biomechanical or clinically significant difference in outcomes remains unclear and requires further investigation with longer-term follow-up.

Conclusions

In this preliminary short-term (1-year) analysis, inverse probability weighting-adjusted rates of overall distal failure or sacroiliac joint dysfunction did not differ significantly between PFFI and standard threaded S2AI implants. PFFI was associated with a lower adjusted rate of radiographic pelvic screw loosening; however, this secondary finding was based on 6 events and may be influenced by systematic differences in implant size and residual confounding; thus, its clinical significance remains uncertain. Larger prospective studies using standardized CT assessment and longer follow-up (≥2 years) are needed.

Supplementary material

Footnotes

  • Funding The authors received no financial support for the research, authorship, and/or publication of this article.

  • Declaration of Conflicting Interests C.S. and S.L. receive consulting fees from SI Bone.

  • Disclosures Charles Sansur reports royalties/licenses from Globus; consulting fees from SI Bone, Globus, and Medtronic; and serves as the treasurer of the AANS/CNS Spine Section. Julio Jauregui reports consulting fees from Globus. Steven Ludwig reports grants or contracts from Baxter and fellowship support from the AOA and AO Spine (all paid to institution); royalties or licenses from ATEC, Atlas Spine, Globus, K2M/Stryker, and SI Bone; consulting fees from Globus, K2M/Stryker, Atlas Spine, ATEC, MDC, ASIP, and SI Bone; payment or honoraria from Globus; and a leadership role (president) of CSRS. Timothy Chryssikos reports consulting fees from GE Healthcare.

  • Ethics Considerations This study was approved by the local Institutional Review Board with consent exemption under 45 CFR 46.101(b).

  • Data Availability Statement Data available upon reasonable request.

References

  1. 1.
    ↵
    Lebwohl NH , Cunningham BW , Dmitriev A , et al . Biomechanical comparison of lumbosacral fixation techniques in a calf spine model. Spine (Phila Pa 1976). 2002;27(21):2312–2320. 10.1097/00007632-200211010-00003
    Google ScholarOpenURLCrossRef
  2. 2.
    ↵
    McCord DH , Cunningham BW , Shono Y , Myers JJ , McAfee PC . Biomechanical analysis of lumbosacral fixation. Spine (Phila Pa 1976). 1992;17(8 Suppl):S235–S243. 10.1097/00007632-199208001-00004
    Google ScholarOpenURLCrossRef
  3. 3.
    ↵
    Cunningham BW , Sponseller PD , Murgatroyd AA , Kikkawa J , Tortolani PJ . A comprehensive biomechanical analysis of sacral alar iliac fixation: an in vitro human cadaveric model. J Neurosurg Spine. 2019;30(3):367–375. 10.3171/2018.8.SPINE18328
    Google ScholarOpenURLCrossRef
  4. 4.
    ↵
    Shin JK , Lim B-Y , Goh TS , et al . Effect of the screw type (S2-alar-iliac and iliac), screw length, and screw head angle on the risk of screw and adjacent bone failures after a spinopelvic fixation technique: a finite element analysis. PLoS One. 2018;13(8):e0201801. 10.1371/journal.pone.0201801
    Google ScholarOpenURLCrossRef
  5. 5.
    ↵
    Rahmani R , Stegelmann SD , Andreshak T . S2 alar-iliac screws are superior to traditional iliac screws for spinopelvic fixation in adult spinal deformity: a systematic review and meta-analysis. Spine Deform. 2024;12(3):829–842. 10.1007/s43390-024-00834-x
    Google ScholarOpenURLCrossRef
  6. 6.
    ↵
    Hasan MY , Liu G , Wong HK , Tan JH . Postoperative complications of S2AI versus iliac screw in spinopelvic fixation: a meta-analysis and recent trends review. Spine J. 2020;20(6):964–972. 10.1016/j.spinee.2019.11.014
    Google ScholarOpenURLCrossRef
  7. 7.
    ↵
    De la Garza Ramos R , Nakhla J , Sciubba DM , Yassari R . Iliac screw versus S2 alar-iliac screw fixation in adults: a meta-analysis. J Neurosurg. 2019;30(2):253–258. 10.3171/2018.7.SPINE18710
    Google ScholarOpenURLCrossRef
  8. 8.
    ↵
    Kim KT , Chryssikos T , Hentschel M , Crandall K , Ludwig S , Sansur CA . Predictors of postoperative sacroiliac joint dysfunction and distal failure after iliac and S2-alar-iliac spinopelvic fixation. J Neurosurg Spine. 2025;44(1):153–164. 10.3171/2025.6.SPINE25395
    Google ScholarOpenURLCrossRef
  9. 9.
    ↵
    Elder BD , Lakomkin N , Zuckerman SL , et al . Distal junctional failure following pelvic instrumentation in spinal fusion: a novel classification system. Neurosurg Focus. 2025;58(6):E14. 10.3171/2025.3.FOCUS24969
    Google ScholarOpenURLCrossRef
  10. 10.
    ↵
    Alan N , Lee KE , Zhou JJ , et al . Bilateral sacroiliac joint fusion in long constructs using self-harvesting porous S2-alar iliac screws with an integrated tulip: technical considerations and early clinical and radiographic experience. Neurosurg Focus. 2023;55(1):E5. 10.3171/2023.4.FOCUS23149
    Google ScholarOpenURLCrossRef
  11. 11.
    ↵
    Hendrickson NR , Haselhuhn JJ , Odland K , et al . Sacropelvic fixation with porous fusion/fixation screws: a technical note and retrospective review. World Neurosurg. 2024;190:172–180. 10.1016/j.wneu.2024.06.046
    Google ScholarOpenURLCrossRef
  12. 12.
    ↵
    Eastlack RK , Menger RP , Turner JD , Ashcraft KR , Carlton Recking W , Kleck CJ . Spinopelvic fixation using an osseointegrative implant: analysis of postmarket surveillance to determine the failure rate. Int J Spine Surg. 2025;19(3):273–278. 10.14444/8720
    Google ScholarOpenURLCrossRefAbstract/Full TextPubMed
  13. 13.
    ↵
    Zou D , Li W , Xu F , Du G . Use of hounsfield units of S1 body to diagnose osteoporosis in patients with lumbar degenerative diseases. Neurosurg Focus. 2019;46(5):E6. 10.3171/2019.2.FOCUS18614
    Google ScholarOpenURLCrossRef
  14. 14.
    ↵
    Shimizu T , Lenke LG , Cerpa M , et al . A radiographic analysis of lumbar fusion status and instrumentation failure after complex adult spinal deformity surgery with spinopelvic fixation: two-year follow-up from the scoli-risk-1 prospective database. Clin Spine Surg. 2020;33(10):E545–E552. 10.1097/BSD.0000000000001008
    Google ScholarOpenURLCrossRef
  15. 15.
    ↵
    Huijbregts P . Sacroiliac joint dysfunction: evidence-based diagnosis. Orthopaedic Division Review. 2004;(May/June):18–43.
    Google Scholar
  16. 16.
    ↵
    Murata S , Iwasaki H , Takami M , et al . Sacroiliac joint pain should be suspected in early buttock and groin pain after adult spinal deformity surgery: an observational study. Spine Surg Relat Res. 2022;6(5):472–479. 10.22603/ssrr.2021-0196
    Google ScholarOpenURLCrossRef
  17. 17.
    ↵
    Buchanan P , Vodapally S , Lee DW , et al . Successful diagnosis of sacroiliac joint dysfunction. J Pain Res. 2021;14:3135–3143. 10.2147/JPR.S327351
    Google ScholarOpenURLCrossRefPubMed
  18. 18.
    ↵
    Petersen T , Laslett M , Juhl C . Clinical classification in low back pain: best-evidence diagnostic rules based on systematic reviews. BMC Musculoskelet Disord. 2017;18(1):188. 10.1186/s12891-017-1549-6
    Google ScholarOpenURLCrossRef
  19. 19.
    ↵
    Kennedy DJ , Engel A , Kreiner DS , Nampiaparampil D , Duszynski B , MacVicar J . Fluoroscopically guided diagnostic and therapeutic intra-articular sacroiliac joint injections: a systematic review. Pain Med. 2015;16(8):1500–1518. 10.1111/pme.12833
    Google ScholarOpenURLCrossRefPubMed
  20. 20.
    ↵
    Hernan MA , Robins JM . Causal inference: what if. Causal Inference: What If. Chapman & Hall/CRC; 2020.
    Google Scholar
  21. 21.
    ↵
    Chen C , Chen S , Ye Z , Shi X , Ma T , Shardell M . The effect of alcohol intake on brain white matter microstructural integrity: a new causal inference framework for incomplete phenomic data. J Am Stat Assoc. 2026;121(553):181–193. 10.1080/01621459.2025.2542553
    Google ScholarOpenURLCrossRef
  22. 22.
    ↵
    Tang X , Yang R , Qu H , Cai Z , Guo W . Factors associated with spinopelvic fixation mechanical failure after total sacrectomy. Spine (Phila Pa 1976). 2018;43(18):1268–1274. 10.1097/BRS.0000000000002635
    Google ScholarOpenURLCrossRef
  23. 23.
    ↵
    Finger T , Bayerl S , Bertog M , Czabanka M , Woitzik J , Vajkoczy P . Impact of sacropelvic fixation on the development of postoperative sacroiliac joint pain following multilevel stabilization for degenerative spine disease. Clin Neurol Neurosurg. 2016;150:18–22. 10.1016/j.clineuro.2016.08.009
    Google ScholarOpenURLCrossRef
  24. 24.
    ↵
    Eastlack RK , Soroceanu A , Mundis GM , et al . Rates of loosening, failure, and revision of iliac fixation in adult deformity surgery. Spine (Phila Pa 1976). 2022;47(14):986–994. 10.1097/BRS.0000000000004356
    Google ScholarOpenURLCrossRef
  25. 25.
    ↵
    Park SJ , Park JS , Nam Y , Yum TH , Choi YT , Lee CS . Failure types and related factors of spinopelvic fixation after long construct fusion for adult spinal deformity. Neurosurg. 2021;88(3):603–611. 10.1093/neuros/nyaa469
    Google ScholarOpenURLCrossRef
  26. 26.
    ↵
    Martin CT , Holton KJ , Elder BD , et al . Catastrophic acute failure of pelvic fixation in adult spinal deformity requiring revision surgery: a multicenter review of incidence, failure mechanisms, and risk factors. J Neurosurg Spine. 2023;38(1):98–106. 10.3171/2022.6.SPINE211559
    Google ScholarOpenURLCrossRef
  27. 27.
    ↵
    Jones KE , Holton KJ , Ladd B , et al . Pelvic fixation in adult spinal deformity: complications, controversies, and future directions. Global Spine J. 2025;15(3_suppl):123S–134S. 10.1177/21925682241242454
    Google ScholarOpenURLCrossRef
  28. 28.
    ↵
    Shin HK , Park JH , Jeon SR , et al . Sacropelvic fixation for adult deformity surgery comparing iliac screw and sacral 2 alar-iliac screw fixation: systematic review and updated meta-analysis. Neurospine. 2023;20(4):1469–1476. 10.14245/ns.2346654.327
    Google ScholarOpenURLCrossRefPubMed
  29. 29.
    ↵
    Lee NJ , Park PJ , Puvanesarajah V , et al . How common is acute pelvic fixation failure after adult spine surgery? A single-center study of 358 patients. J Neurosurg Spine. 2023;38(1):91–97. 10.3171/2022.7.SPINE22498
    Google ScholarOpenURLCrossRef
  30. 30.
    ↵
    Jankowski PP , Tahmasebpour P , Matthews S , et al . Is more better? Multicenter analysis of the incidence and mechanisms of multiple pelvic fixation failure in adult spinal deformity surgery. Spine J. 2026;26(7):1378–1390. 10.1016/j.spinee.2025.12.013
    Google ScholarOpenURLCrossRef
  31. 31.
    ↵
    Panico M , Chande RD , Lindsey DP , et al . Stability and instrumentation stresses among sacropelvic fixation techniques with novel porous fusion/fixation implants: a finite element study. Int J Spine Surg. 2023;17(4):598–606. 10.14444/8481
    Google ScholarOpenURLCrossRefAbstract/Full TextPubMed
  32. 32.
    ↵
    Panico M , Chande RD , Lindsey DP , et al . The use of triangular implants to enhance sacropelvic fixation: a finite element investigation. Spine J. 2020;20(10):1717–1724. 10.1016/j.spinee.2020.05.552
    Google ScholarOpenURLCrossRef
  33. 33.
    ↵
    Gao Z , Sun X , Chen C , et al . Comparative radiological outcomes and complications of sacral-2-alar iliac screw versus iliac screw for sacropelvic fixation. Eur Spine J. 2021;30(8):2257–2270. 10.1007/s00586-021-06864-7
    Google ScholarOpenURLCrossRefPubMed
  34. 34.
    ↵
    Luo Q , Kim Y-C , Kim K-T , et al . Use of iliac screw associated with more correction of lumbar lordosis than S2-alar-iliac screw for adult spinal deformity. BMC Musculoskelet Disord. 2021;22(1):676. 10.1186/s12891-021-04568-z
    Google ScholarOpenURLCrossRef
  35. 35.
    ↵
    MacBarb RF , Lindsey DP , Woods SA , Lalor PA , Gundanna MI , Yerby SA . Fortifying the bone-implant interface part 2: an in vivo evaluation of 3D-printed and TPS-coated triangular implants. Int J Spine Surg. 2017;11(3):16. 10.14444/4016
    Google ScholarOpenURLCrossRef
  36. 36.
    ↵
    Arena JD , Ghenbot Y , Dagli MM , et al . Incidence and outcomes of spinopelvic fixation loosening following adult spinal deformity surgery. Spine Deform. 2026;14(5):1661–1669. 10.1007/s43390-026-01281-6
    Google ScholarOpenURLCrossRef
  37. 37.
    ↵
    Ivanov AA , Kiapour A , Ebraheim NA , Goel V . Lumbar fusion leads to increases in angular motion and stress across sacroiliac joint: a finite element study. Spine (Phila Pa 1976). 2009;34(5):E162–E169. 10.1097/BRS.0b013e3181978ea3
    Google ScholarOpenURLCrossRef
  38. 38.
    ↵
    Hyun SJ , Jung JM , Kim KJ , Jahng TA . Durability and failure types of S2-alar-iliac screws: an analysis of 312 consecutive screws. Oper Neurosurg. 2020;20(1):91–97. 10.1093/ons/opaa251
    Google ScholarOpenURLCrossRef
  39. 39.
    ↵
    Manzetti M , Ruffilli A , Barile F , et al . Sacroiliac joint degeneration and pain after spinal arthrodesis: a systematic review. Clin Spine Surg. 2023;36(4):169–182. 10.1097/BSD.0000000000001341
    Google ScholarOpenURLCrossRef
  40. 40.
    ↵
    Sohn S , Park TH , Chung CK , et al . Biomechanical characterization of three iliac screw fixation techniques: a finite element study. J Clin Neurosci. 2018;52:109–114. 10.1016/j.jocn.2018.03.002
    Google ScholarOpenURLCrossRef
  41. 41.
    ↵
    de Andrada Pereira B , Wangsawatwong P , Lehrman JN , et al . Biomechanics of a laterally placed sacroiliac joint fusion device supplemental to S2 alar-iliac fixation in a long-segment adult spinal deformity construct: a cadaveric study of stability and strain distribution. J Neurosurg Spine. 2022;36(1):42–52. 10.3171/2021.3.SPINE202175
    Google ScholarOpenURLCrossRef
View Abstract
Loading
Loading
PDF
Back to top

In this Issue

International Journal of Spine Surgery

Vol. 20, Issue 4

1 Aug 2026

Table of Contents Index by author

Jump to Section

Keywords

PFFI, S2AI, s2-alar-iliac, spinopelvic fixation, distal failure, sacroiliac joint dysfunction