Comparative efficacy of autologous iliac bone versus allogeneic demineralized bone matrix in lateral lumbar interbody fusion using a novel minimally invasive iliac bone retrieval tool: a self-controlled study

Abstract

Objective

This self-controlled study evaluates the fusion efficacy of autologous iliac bone harvested via a novel minimally invasive tool versus allogeneic demineralized bone matrix (DBM) in lateral lumbar interbody fusion (LLIF), while assessing the safety profile of the retrieval tool.

Methods

This study was a prospective clinical controlled study. Patients’ basic information was recorded, including the age, gender, body mass index and etc. Visual Analog Scale (VAS) and Oswestry Disability Index (ODI) were used to evaluate clinical efficacy. Three parameters were measured, including intervertebral space height, intervertebral foraminal height and lumbar lordosis angle, before surgery and 3 days/24 months after surgery. Postoperative CT scan was used to compare the difference in interbody fusion between autologous iliac bone and allogeneic bone DBM at 6 and 24 months after surgery.

Results

30 patients followed up for more than 24 months were included. The preoperative VAS for lower back and leg pain was 5.00 ± 0.87, and the preoperative ODI was 48.37 ± 8.53. The VAS for lower back and leg pain was 0.87 ± 0.63, and the ODI was 12.30 ± 2.58 at 24 months after surgery. The VAS and ODI in each postoperative stage were significantly improved compared to the preoperative (P < 0.05). The VAS of the bone extraction area was 1 (0,2) after surgery, and the pain disappeared approximately 2 days after surgery. At 6 months postoperatively, the fusion rate was 66.7% (20/30) for autologous iliac bone versus 30.0% (9/30) for allogeneic DBM (P < 0.05). the fusion rate on the autogenous iliac bone side (96.7%) was significantly higher than that on the DBM side (70.0%) at 24 months after surgery (P < 0.05).

Conclusions

The novel retrieval tool enables safe, minimally invasive harvest of autologous iliac bone, which achieved significantly higher fusion rates than allogeneic DBM in LLIF.

Introduction

Lumbar fusion surgery has been improved in fusion site selection and surgical approach. The optimal choice of fusion site has changed from lumbar posterolateral fusion and transverse process fusion to lumbar interbody fusion [1–4]. Lumbar interbody fusion has become the mainstream lumbar fusion method due to its mechanical advantages. Lumbar interbody fusion surgery originated from open anterior lumbar interbody fusion in the early 20th century. Subsequently, scholars have proposed and practiced posterior lumbar interbody fusion (PLIF), transforaminal lumbar interbody fusion (TLIF), minimally invasive transforaminal lumbar interbody fusion (MIS-TLIF), and now the widely popular minimally invasive lateral lumbar interbody fusion (LLIF) [5–9]. The autologous bone harvested during PLIF, TLIF, and MIS-TLIF, surgeries typically suffices for the autologous bone required for interbody fusion. Oblique lumbar interbody fusion (OLIF) and extreme lateral interbody fusion (XLIF) are two minimally invasive lateral lumbar interbody fusion techniques that approach through the retroperitoneal space, directly reaching the anterolateral aspect of the disc without bone removal, and therefore do not harvest autologous bone for interbody grafting during surgery [8, 10]. Early solid interbody fusion is critical to achieving favorable clinical outcomes in LLIF procedures. The type of interbody fusion material used in LLIF to promote fusion has been research hotspot [11–14].

Currently, lumbar lateral interbody fusion surgery commonly uses polyetheretherketone (PEEK) cages, with the hollow part of the cage filled with bone graft material [15–17]. Autologous bone is regarded as the gold standard for bone graft materials, with autologous iliac bone being the first choice [18, 19]. However, open iliac bone harvesting is associated with complications such as donor site infection, pain, and iliac bone fractures, which are contrary to the principles of minimally invasive surgery [20–22]. Currently, a widely used alternative to autologous bone grafts in lumbar lateral interbody fusion surgery involves filling the hollow part of the cage with allogeneic demineralized bone matrix (DBM) [11]. However, allogeneic DBM lacks osteoinductive properties and has a low early fusion rate [12]. Related basic and clinical research showed that various bone growth factors, such as bone morphogenetic proteins, vascular endothelial growth factor, and colony-stimulating factors, exhibited significant osteoinductive properties during interbody fusion processes [23, 24]. These factors can act independently or synergistically. Our prior study also demonstrated that applying rhBMP-2 in rabbit lumbar lateral interbody fusion surgery achieved a higher fusion rate compared to using allograft bone alone [25]. However, these commercial bone growth factors are costly and they pose other potential adverse effects.

Considering the current clinical application of fusion materials in lumbar lateral interbody fusion surgery, a specialized minimally invasive iliac bone harvesting tool was designed to be used during lumbar lateral interbody fusion surgery. To address donor-site morbidity associated with traditional iliac bone harvesting, we developed a novel minimally invasive tool. Using a self-controlled design, this study investigates whether autologous iliac bone obtained via this tool achieves superior fusion rates compared to allogeneic DBM in LLIF.

Methods

Inclusion and exclusion criteria

This study is a prospective clinical controlled study comparing the interbody fusion differences between autologous iliac bone and allograft DBM in lumbar lateral interbody fusion surgery. This study was approved by the Hospital Ethics Committee (20377750D). Inclusion criteria: Lumbar degenerative diseases such as lumbar spinal stenosis, lumbar instability, mild lumbar spondylolisthesis that were ineffective after 3 months of strict conservative treatment, undergoing single-segment XOLIF surgery [26]. Basic information such as age, gender, and body mass index of the patients was recorded. Exclusion criteria: (1) Lumbar infection, tumor, trauma; (2) Previous surgery in the retroperitoneal space; (3) Incomplete follow-up data.

Surgical procedure

Following general anesthesia, the patient was positioned in a right lateral decubitus position with hips and knees flexed. The XOLIF surgical method was employed to treat the intervertebral space, revealing the bony endplate of the surgical segment [26]. An additional 1-cm surgical incision was made 2 cm lateral to the anterior superior iliac spine and 2 cm caudal. Using the novel minimally invasive iliac bone harvesting tool, a 2 × 1 × 1 cm³ iliac bone block was obtained for later use. An appropriately sized PEEK cage was selected. The cranial hole (right side) was filled with autologous iliac bone, and the caudal hole (left side) was filled with allogeneic DBM. The intervertebral space was implanted with a PEEK cage filled with autologous iliac bone and allogeneic DBM. Intraoperative confirmation of the cage position was satisfactory in both anteroposterior and lateral views. Patients were evaluated for osteoporosis preoperatively. To control for fixation-related confounding, patients with osteoporosis (DEXA T-score ≤ -2.5) underwent posterior bilateral percutaneous pedicle screw fixation, while non-osteoporotic patients received lateral interbody fixation. Fixation method was included as a covariate in statistical analyses. The study used a self-controlled study method to reduce the impact of different internal fixation methods on the study. After thorough hemostasis, the incision was sutured without placing a drain. Figure 1 showed the specific process of using minimally invasive iliac bone harvested for interbody fusion in this study.

Fig. 1.

The specific process of minimally invasive iliac bone extraction for interbody fusion in this study. (A) Take the skin entry point at 2 cm on the lateral and caudal side of the anterior superior iliac spine. (B) After drilling through the inner and outer cortical layers of the iliac bone with a circular saw, remove the bone retrieval tool. (C) Remove the iliac bone block by tapping the tail end of the inner sleeve. (D) A set of minimally invasive iliac bone retrieval tools. (E) Double hole PEEK cage with autologous iliac bone filling in the cephalic foramen (right side) and allogeneic bone powder filling in the caudal foramen (left side). (F) The incision of lumbar lateral interbody fusion and minimally invasive iliac bone extraction

Clinical efficacy

The recorded data included intraoperative blood loss, surgery duration, surgical segments, and perioperative complications. The Visual Analog Scale (VAS) was used to evaluate lumbar and leg pain preoperatively, three days postoperatively, and during each follow-up period; and was also used to assess pain at the iliac bone harvested site postoperatively. The Oswestry Disability Index (ODI) was used to assess postoperative functional improvement.

Radiological assessment

The intervertebral space height (average distance between the anterior and posterior edges of adjacent endplates), foraminal height (maximum distance between the upper and lower edges of the foramen), and lumbar lordosis angle (angle between the upper endplate of L1 and the upper endplate of S1) were measured preoperatively, 3 days postoperatively, and 24 months postoperatively. Fusion assessment was performed using CT scans at 6 and 24 months postoperatively. Two blinded spine surgeons independently applied the modified Brantigan criteria [27], with inter-rater reliability calculated (κ = 0.85) ( Fig. 2). The differences in fusion of cephalad and caudal foramen fillers were compared by sagittal and coronal reconstruction of CT. Brantigan grading criteria: Grade 1 - no connection between the endplates, gap collapse, graft absorption; Grade 2 - no connection between the endplates, increased bone mass compared to immediate postoperative; Grade 3 - partial (50%) connection between the endplates, still many translucent lines; Grade 4 - good fusion between the endplates, few translucent lines; Grade 5 - complete fusion between the endplates. Solid interbody fusion is defined as ≥ Grade 4.

Fig. 2.

CT fusion status at 3 days, 6 months, and 24 months after lumbar spine 3-4XOLIF surgery. (A, B, C) CT axial images at 3 days, 6 months, and 24 months after surgery; (D, E, F) CT coronal images at 3 days, 6 months, and 24 months after surgery; (G, H, I) Sagittal CT scan of the decalcified bone matrix side of allograft bone at 3 days, 6 months, and 24 months after surgery; (J, K, L) Sagittal CT scan of autogenous iliac bone at 3 days, 6 months, and 24 months after surgery

Statistical analysis

Statistical analysis was performed using SPSS 23.0 statistical software (IBM, Armonk, NY, USA). If the measurement data conformed to a normal distribution, it was described using mean ± standard deviation; otherwise, it was expressed as median (25th percentile, 75th percentile). Categorical data were described using percentages. One-way repeated measures ANOVA was used to compare the differences in VAS and ODI at different time points; the chi-square test was used to compare the fusion differences between autologous iliac bone and allograft bone, with P < 0.05 indicating statistical significance.

Results

Patient demographics

30 patients who were followed up for more than 24 months were included. The basic information of the 30 patients included in the study (11 males, 19 females) was presented in Table 1. The patients’ mean age was 61.57 ± 6.85 years. Among these cases, 7 involved the L3-4 segment and 23 involved the L4-5 segment; 9 patients received lateral interbody fixation, and 21 received posterior percutaneous pedicle screw fixation. The average intraoperative blood loss was 62.00 ± 21.88 ml, and the average surgical time was 104.03 ± 21.16 min. No surgical complications occurred in this group during the perioperative period. The average follow-up duration for this group of patients was 27.36 ± 2.39 months. The harvested autologous iliac bone volume averaged 2.0 cm³ per patient, sufficient to fill the cranial cage hole. No correlation was observed between graft volume and fusion rate.

Table 1.

Basic information of surgical patients

Values
Age (year)	61.57 ± 6.85
Gender (male/female)	11/19
Surgical time (minutes)	104.03 ± 21.16
Intraoperative bleeding volume (ml)	62.00 ± 21.88
Follow up time (month)	27.36 ± 2.39
Surgical segment
L3-4	7 (23.3%)
L4-5	23 (76.7%)
Internal fixation method
lateral interbody fixation	9 (30.0%)
posterior bilateral percutaneous pedicle screw fixation	21 (70.0%)

Clinical efficacy

The preoperative and postoperative VAS scores for lumbar and leg pain, and ODI for this group of patients were shown in Table 2. The mean ± standard deviation of the preoperative VAS for lumbar and leg pain was 5.00 ± 0.87, and the preoperative ODI was 48.37 ± 8.53. At 24 months postoperatively, the mean ± standard deviation of the VAS for lumbar and leg pain was 0.87 ± 0.63, and the ODI was 12.30 ± 2.58. The VAS and ODI for lumbar and leg pain at all postoperative stages were significantly improved compared to preoperative values (P < 0.05). The mean VAS and ODI scores for lumbar and leg pain decreased with follow-up time. The VAS and ODI at 3 months, 6 months, and 24 months postoperatively were significantly lower than those at 3 days postoperatively (P < 0.05). Donor-site complications included transient pain (VAS 1 [0–2]), resolving within 48 h, and no instances of infection, hematoma, or fracture.

Table 2.

Comparison of VAS for lower back and leg pain and ODI before and after surgery

Values	Pre-op	3 days	3 months	6 months	24 months	P (2-Tailed)
VAS	5.00 ± 0.87a	2.30 ± 0.99b	1.20 ± 0.66c	0.96 ± 0.61c	0.87 ± 0.63c	< 0.001$
ODI (100%)	48.37 ± 8.53a	19.67 ± 3.47b	13.20 ± 3.19c	12.93 ± 3.24c	12.30 ± 2.58c	< 0.001$

VAS indicates Visual Analog Scale; ODI, Oswestry Disability Index; ^$, One-way repeated measures ANOVA; ^#, the chi-square test; Multiple comparisons of each variable at different time points using the Bonferroni method, and at least 1 identical subscript letter denoted no significant difference from each other at the 0.05 level

Radiographic evaluation

Table 3 illustrated the changes in intervertebral space height, foraminal height, and lumbar lordosis angle preoperatively, 3 days postoperatively, and 24 months postoperatively. The preoperative intervertebral space height was 7.24 ± 0.90 mm, which significantly increased to 10.34 ± 1.16 mm at 3 days postoperatively and 9.90 ± 1.02 mm at 24 months postoperatively. The foraminal height increased significantly from 14.92 ± 1.89 mm preoperatively to 20.17 ± 1.83 mm at 3 days postoperatively and 19.90 ± 1.77 mm at 24 months postoperatively. The intervertebral space height and foraminal height at 24 months postoperatively decreased by 0.44 mm and 0.27 mm, respectively, compared to 3 days postoperatively; although these differences were statistically significant, they were not clinically significant. The lumbar lordosis angles were 35.00° ± 5.54 preoperatively, 35.80° ± 4.63 at 3 days postoperatively, and 35.43°±4.10 at 24 months postoperatively, with no statistically significant differences among them. Six months postoperatively, the solid fusion rate on the autologous iliac bone side was 66.7% (20/30) compared to 30.0% (9/30) on the allogeneic DBM side; the solid fusion rate on the autologous iliac bone side was significantly higher than that on the allogeneic DBM side (P < 0.05). Twenty-four months postoperatively, the solid fusion rate was 96.7% (29/30) on the autologous iliac bone side and 70.0% (21/30) on the allogeneic DBM side; the solid fusion rate on the autologous iliac bone side was significantly higher than that on the allogeneic DBM side (P < 0.05).

Table 3.

Comparison of intervertebral height, foraminal height and lumbar Lordotic angle before and after surgery

Values	Pre-op	3 days	24 months	P (2-Tailed)
Intervertebral space height(mm)	7.24 ± 090a	10.34 ± 1.16b	9.90 ± 1.02c	< 0.001$
foraminal height(mm)	14.92 ± 1.89a	20.17 ± 1.83b	19.90 ± 1.77c	< 0.001$
lumbar lordosis angle(°)	35.00 ± 5.54a	35.80 ± 4.63a	35.43 ± 4.10a	0.249$

^$ indicates One-way repeated measures ANOVA; Multiple comparisons of each variable at different time points using the Bonferroni method, and at least 1 identical subscript letter denoted no significant difference from each other at the 0.05 level

Discussion

The LLIF technique accesses the side of the intervertebral space through the retroperitoneal space, a notable feature being the absence of autologous bone harvested during the procedure. Spine surgeons often use PEEK cages for interbody fusion, with the hollow part filled with bone graft material. Clinical studies have confirmed that using allogeneic bone alone results in lower early fusion rates due to the lack of osteoinductive properties. Basic research has shown that combining allogeneic bone with bone growth factors can enhance lumbar interbody fusion rates [28, 29]. rhBMP-2 is a clinically available bone growth factor widely used [30]. However, the clinical application of rhBMP-2 is associated with immune reactions, nerve irritation, and increased hospitalization costs [31]. Although autologous iliac bone offers advantages such as no immune rejection and low cost, current harvesting methods are associated with complications such as pain, bleeding, and infection at the donor site [32]. Our team has innovatively developed a minimally invasive iliac bone harvesting tool to reduce donor site complications while safely and efficiently obtaining the iliac bone needed to fill the PEEK cage. In this study, the cage was not completely filled with autologous iliac bone; instead, a two-hole cage was used, with one hole filled with minimally invasive harvested iliac bone and the other with allogeneic DBM. This study compared the fusion differences between these two areas using self-control.

In this study, the 30 patients showed significant improvement in VAS and ODI scores for lumbar and leg pain at each postoperative follow-up compared to preoperative values, with satisfactory clinical efficacy similar to previous studies. The modified XOLIF technique used in this study did not result in any perioperative complications, which is related to the technique’s splitting of the anterior edge of the psoas major tendon under direct vision, combining the advantages of XLIF and OLIF. Postoperative pain at the iliac bone harvesting site was not significant, and for those who experienced postoperative pain, it generally resolved shortly after surgery, highlighting the minimally invasive advantage of this bone harvesting method. Literature reports frequently describe iliac bone harvesting methods involving the removal of three-sided cortical bone blocks from the posterior iliac crest, which are associated with significant bleeding, damage, and large residual bone defects [33, 34]. The minimally invasive iliac bone harvesting method used in this study preserved the cortical bone of the iliac crest edges, removing only the required amount of two cortical iliac bone blocks, making it more precise and minimally invasive, while adequately protecting the surrounding soft tissues. The tool’s efficacy is supported by precise cortical preservation (Fig. 1D) and minimal donor-site morbidity, contrasting with literature reports of 15–30% complication rates with open harvesting [20, 33].

In this study, the foraminal height and intervertebral space height were significantly improved at 3 days and 24 months postoperatively compared to preoperative values; no significant cage subsidence or intervertebral space collapse was observed during the follow-up period. Research indicates that risk factors for cage subsidence and intervertebral space collapse include high body mass index, osteoporosis, improper cage size selection, inappropriate fixation methods, and intraoperative endplate damage [35–37]. The absence of significant cage subsidence and the maintenance of foraminal height and intervertebral space height in this study can be attributed to the following: (1) The use of XOLIF did not involve sharp curettes for space treatment, thus adequately protecting the bony endplates; (2) Appropriate selection of cage size; (3) Determination of fixation method based on the patient’s bone quality; (4) The autologous iliac bone side achieved satisfactory fusion in the early postoperative period. The changes in lumbar lordosis angle before and after surgery were minimal in this study. Park et al. [38] have demonstrated that LLIF can help improve lumbar sagittal curvature and reconstruct lumbar lordosis. We believe the minimal changes in lumbar lordosis angle before and after surgery in our patients are related to the conditions of the surgical segments, as there was no significant local kyphosis in this group. Additionally, this study only included single-segment surgeries, which does not fully reflect the advantages of LLIF in reconstructing spinal sagittal balance. Our findings demonstrate significantly higher fusion rates with autologous iliac bone compared to allogeneic DBM (P < 0.05), consistent with its osteoinductive properties [18, 39].

This study has limitations: (1) single-center design with a modest sample size (n = 30), (2) variable fixation methods, though statistically adjusted, and (3) absence of direct comparison to traditional iliac bone harvesting techniques. Future multi-center trials with standardized fixation are warranted.

Through a self-controlled study, we confirmed that using our innovative minimally invasive iliac bone harvesting tool for lumbar lateral interbody fusion surgery achieves good clinical outcomes and solid interbody fusion without increasing donor site complications. The self-controlled study design is a feature of this study, but it only involved implanting autologous iliac bone in one side hole of the cage, not both. The results showed that implanting autologous iliac bone in one side hole achieved good interbody fusion and maintained effective indirect decompression at the end of the follow-up period. We speculate that if autologous iliac bone blocks were implanted in both side holes, solid interbody fusion would be achieved early postoperatively on both sides. Future clinical applications could involve using the same iliac bone harvesting skin incision to reposition the minimally invasive iliac bone harvesting tool to obtain a second double-cortical bone block, filling both sides of the PEEK cage with iliac bone blocks.

Conclusions

The novel retrieval tool enables safe, minimally invasive harvest of autologous iliac bone, which achieved significantly higher fusion rates than allogeneic DBM in LLIF. While long-term stability was observed, early fusion (6-month) rates remained superior for autograft, supporting its osteogenic advantage.

Abbreviations

LLIF

Lateral lumbar interbody fusion

VAS

Visual Analog Scale

ODI

Oswestry Disability Index

PLIF

Posterior lumbar interbody fusion

TLIF

Transforaminal lumbar interbody fusion

MIS-TLIF

Minimally invasive transforaminal lumbar interbody fusion

OLIF

Oblique lumbar interbody fusion

XLIF

Extreme lateral interbody fusion

Author contributions

Jiaqi Li & Shaorong Li: Conceptualization, Data curation, Writing - review & editing, Writing - original draft. Shuowen Zhang: Data curation. Lin Liu: Patient follow-up. Weijian Wang: Patient follow-up. Han Wu: Writing– review. Qiang Yang: Writing– review. Wei Zhang: Conceptualization, Funding acquisition.All authors reviewed the manuscript.

Funding

This work was supported by Bureau of Science and Technology of Hebei Province (20377750D), Hebei Medical Science Research Project (20221185) and 2023 Hebei Medical University Clinical Medical Innovation Research Team Support Program.

Data availability

Data is provided within the manuscript.

Declarations

Ethics approval and consent to participate

This study was a prospective study and achieved the agreement of the hospital ethics committee. All patients’ data collection was informed in advance.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.