Lateral Lumbar Interbody Fusion With rhBMP-2 can Achieve High Fusion Rates in Adult Spine Deformity Surgeries

Abstract

Study design

An ambispective, observational study of the prospective, institutional adult spine deformity (ASD) database.

Objectives

The study investigates the clinical and radiographic fusion rates with lateral interbody approach and rhBMP-2 in multiple-level lumbar fusion in the ASD population. Previous studies have reported over 10% pseudoarthrosis rate with multiple segment fusions. Lateral lumbar interbody fusion (LLIF) allows multiple-level, less-invasive access to the anterior lumbar spine. We hypothesized that fusion rates with lateral approach with rhBMP-2 use are superior to the published data on lumbar fusion in ASD patients.

Methods

The institutional ASD database was searched to identify eligible patients with two or more levels of LLIF (T12-L5), >4 levels of posterior instrumentations and >2 years of follow-up between the years 2010 and 2018. Antero-posterior and lateral 36-inch standing radiographs for each patient and computed tomography scans in select patients were studied to ascertain fusion status and patients were divided into two groups based on fusion status.

Results

The study included 179 patients with a mean age of 65.3 years and 74% female patients. The median number of interbody fusions was performed at 3 (IQR 3-4) levels. The mean follow-up duration was 4.4 years (SD = 1.9). 169 patients (94.5%) had successful arthrodesis, while 10 patients (5.5%) had radiological pseudoarthrosis at one level. Of the 10 patients, 8 (4.4%) were either clinically asymptomatic or had manageable back pain. Two patients (1.1%) required revision surgery for symptomatic pseudoarthrosis.

Conclusion

This is the largest known series of ASD patients investigating the fusion rates with multiple-level LLIFs. LLIF along with rhBMP-2 can achieve high fusion success across interbody fusion levels in multi-segmental ASD surgeries.

Keywords

adult spinal deformity degenerative scoliosis lateral lumbar interbody fusion lumbar fusion rate pseudoarthrosis

Introduction

Spinal fusion for degenerative spinal conditions has been shown to improve health-related quality of life.¹ With the advent of improved spinal instrumentation and interbody devices, fusion rates have improved, but the optimal choice for graft material remains unclear. Iliac crest autologous bone graft (ICBG) is considered the gold-standard graft material, defined during the era of posterior-alone fusions.¹ However, concerns related to donor-site morbidity increased blood loss and operative time, and limited graft quantity compelled physicians to explore alternatives to the ICBG.² Since the U.S Food and Drug Administration approved rhBMP-2 for single-level ALIFs in 2002,³ rhBMP-2 has been used as an alternative to ICBG and, as a result, its use in spinal surgery has increased substantially. Reported fusion rates with rhBMP-2 are equivalent to ICBG (94.1% with rhBMP-2 vs 89.5% with ICBG) in single-level transforaminal lumbar interbody fusion (TLIF) or posterior lumbar interbody fusion (PLIF)⁴ and posterolateral lumbar fusions (PLF).⁵ While the initial study recommended 20 mg rhBMP-2 use per level,³ dosing recommendations remain inconsistent throughout the literature. Alternative biologic agents (e.g., Silicate calcium phosphate, allogenic morphogenic protein) have demonstrated modest success in comparison to the rhBMP-2.^6,7,9

Superior fusions with rhBMP-2 were initially reported in short-segment constructs, but the fusion rate varies with the surgical approach and the choice of bone-graft material.⁸ The addition of interbody spacers has increased the fusion success (89.5%-91%) as compared to posterior-alone fusion.^4,7 Similarly, the fusion rates with ALIFs and rhBMP-2 are higher than with TLIF or PLIF procedure and rhBMP-2 (97.8% with ALIF vs 95.7% with TLIF or PLIF vs 93.6% with PLF).⁸ In contrast to the short-segment constructs, fusion rates are reportedly lower in the long segment fusion constructs performed in adult spinal deformity (ASD) surgery; initial studies reported 85-90% fusion rates with PLF.^10-12 Kim et al¹³ reported an 83% fusion rate with PLF in the ASD population, utilizing local autogenous bone graft.¹³ There is growing evidence to suggest that rhBMP-2 increases the fusion rate in PLF for ASD surgeries. A recent study reported a 95% arthrodesis rate with rhBMP-2 as compared to 72% with ICBG.¹⁴

In recent years, lateral lumbar interbody fusion (LLIF) has been increasingly performed in ASD surgeries; potential benefits described are improved indirect decompression, higher segmental lordosis correction and improved sagittal alignment.^15,16 Furthermore, multiple LLIFs may be a less-morbid alternative to formal osteotomy procedures for sagittal balance correction.¹⁷ To the best of our knowledge, the arthrodesis rate utilizing rhBMP-2 and multiple-level LLIF in ASD surgeries has not been accurately described. The primary objective of our study was to report on the clinical and radiographic interbody fusion rate with multiple level LLIF using rhBMP-2 in ASD surgeries.

Methods

Study Design

An ambispective observation review of the institutional, prospective ASD database was conducted to identify patients treated for degenerative spinal deformity. Patients treated with two or more levels of LLIFs (between T12-L5 motion segments), >4 levels of posterior instrumentation and a minimum 2-year follow-up were included. Patients with stand-alone cages, single-level interbody fusion, smaller constructs (<4 levels), revision interbody fusion and inadequate follow-ups were excluded from the study. Data were extracted from hospital charts and diagnostic images of patients operated on between 2010-and 2018. A consent was completed before the procedure and Institutional review board approval was obtained before the start of the study. Patients were operated on in a lateral position using a lateral, retroperitoneal ante-psoas or trans-psoas approach. Multiple level LLIFs procedures were performed, and all patients were staged for posterior instrumentation within a week. Most patients also had L5-S1 interbody fusion and either ALIF or TLIF was performed at this level. rhBMP-2 was the only biologic agent used for interbody fusion and was strictly used within the interbody spacer devices; rhBMP-2 was not used in the posterolateral gutters to achieve PLF in primary surgeries. A low dose (3-4 mg) of rhBMP-2 was used at each LLIF level. A larger cantilever force due to long segment fusion alters the biomechanics and fusion success at the lumbosacral junction. Hence, pseudarthrosis at the L5-S1 level was not counted as an event in this analysis.

Radiographic Evaluation

Plain coronal and sagittal 36-inch standing radiographs were obtained pre-procedure and during follow-up. Computed tomography (CT) scans were obtained in select patients with suspicion of pseudoarthrosis or implant-related complications. Moreover, abdominal, and pelvic CT scans obtained for unrelated conditions during the follow-up period were available for review in several patients. All imaging studies obtained within the province of Alberta were available for review, ensuring high imaging capture rates. The upper instrumented vertebra (UIV) was categorized into three groups: upper thoracic (T1-T9), lower thoracic (T10-T12) and upper lumbar (L1-L3). Similarly, the lower instrumented vertebra (LIV) was categorized into three groups: lower lumbar (L4-6), sacrum (S1) and pelvis (Ilium). Coronal and sagittal alignment parameters [Coronal cobb angles, lumbar lordosis (LL), thoracic kyphosis (TK), sagittal vertical axis (SVA), pelvic incidence (PI), pelvic tilt (PT), sacral slope (SS), PI-LL mismatch] were recorded on pre-and postoperative radiographs. Post-operative changes in alignment parameters were analyzed separately to measure the difference between the groups.

Patients with clinical and radiological findings of pseudoarthrosis requiring re-operation were defined as ‘clinically significant’ pseudoarthrosis. Several radiographic characteristics were identified from the existing literature to develop a criterion for the radiological diagnosis of pseudoarthrosis. Two fellowship-trained spine surgeons reviewed images of patients suspected of radiological pseudoarthrosis. The presence of one or more findings, as mentioned below, was suspected of radiological pseudoarthrosis (Table 1). These patients with radiological pseudoarthrosis were either clinically asymptomatic or were unwilling for re-operation. Furthermore, patients were subdivided into two groups (fused and pseudoarthrosis) based on their post-operative radiological findings.

Table 1.

Baseline Characteristics of the Overall Study Population.

Variables		Overall n = 179 (%)
Age (mean, SD)		65.39 (8.3)
Sex
	Males, n (%)	46 (25.7)
	Females, n (%)	133 (74.3)
Upper instrumented vertebra, n(%)
	T1-T9	25 (13.97)
	T10-T12	118 (65.92)
	L1-L3	36 (20.11)
Lower instrumented vertebra, n (%)
	L4-L6	10 (5.59)
	S1	81 (45.25)
	Ilium	88 (49.16)
Levels of instrumentations (median, IQR)		8 (7-9)
Interbody fusion levels (median, IQR)		3 (3-4)
Osteotomies	Posterior column osteotomy	30(16.7)
	Pedicle subtraction osteotomy	8(4.4)
Follow-up duration (years) (mean, SD)		4.4 (1.9)
Revision, n (%)		71 (39.6)
	L5-S1 pseudoarthrosis	18 (25.35)
	Proximal junctional kyphosis	16 (22.5)
	Proximal junctional failure	21 (29.5)
	Rod fracture	5 (7)
	Others (Adjacent level degeneration; ASD, pseudoarthrosis of thoracolumbar fusion mass)	10(14)
	Lateral lumbar interbody fusion Pseudoarthrosis	2 (2.8)
Follow up CT scan, n (%)		120 (67)

Radiographic findings suggesting pseudoarthrosis:

1. Inadequate consolidation of the disc space without bridging trabecular bone at the margins of the spacer or failure of graft incorporation into the vertebral endplates.^18,19

2. Loss of fixation such as implant fracture, or a lucent halo around pedicle screws

3. Progression of radiographic deformity.

4. Progressive spacer device subsidence and spacer migration.¹³

CT findings:

1. Absence of bone trabeculations across fusion levels

2. Bone lucency at graft-vertebral body junction.

3. Halo around pedicle screws or rod fractures,

4. Progressive spacer migration (rotation or translation)^20-22(Figures 1-3)

Figure 1.

(Patient 5): 71 years female treated with L2-5 LLIF and T11-L5 posterior instrumentation. Sagittal computed tomography images (left) show inadequate consolidation and fragmented interbody fusion mass at the L4-5 level. Coronal images (right) also show an absence of bridging trabeculations across disc space. The patient was, however, asymptomatic at the final follow-up. (Table 3).

Figure 2.

(Patient 8): 72 years male underwent L1-5 LLIF and T10-S1 posterior instrumentation. A Follow-up computed tomography image obtained for workup of colorectal carcinoma 4 years after the surgery shows similar findings of inadequate bridging trabeculations across disc space and fragmented fusion mass at the L4-5 level. Bridging bone trabeculation can be noted at the adjacent L2-3 and L3-4 levels.

Figure 3.

(Patient 6): 72 years female underwent L1-5 LLIF and T10-S1 posterior instrumentation (left). Radiographs obtained at 3 years follow-up shows right rod fracture at L1-2 level (arrow). Although adequate consolidation across the disc space could be observed, findings were suspicious for radiological pseudoarthrosis. Significant subsidence was noted at the L4-5 level resulting in several foraminal narrowing and radiculopathy. 72 years Male (Patient 7) was treated with L2-5 LLIF and L2-S1 posterior instrumentation (right). Radiographs obtained 3 years after surgery demonstrate lucency in the anterior disc space at the L2-3 level and halo around the left L2 screw (arrow). The patient also had adjacent segment degeneration of L1-2 disc-space.

Statistical Analysis

Baseline characteristics and imaging features for the overall population were reported using descriptive statistics. Cases were binarized into fused vs pseudarthrosis. Differences in baseline characteristics of patients with successful fusion vs pseudarthrosis were assessed with Fisher’s exact test (categorical variables) and Wilcoxon rank-sum test (continuous variables). Two-sided P < .05 was considered statistically significant. Statistical analysis was performed with Stata/MP 16.1 (StataCorp LP).

Results

Demographic Data

Records of 345 patients were available in the institutional ASD database, and 179 eligible patients were identified after applying inclusion criteria (112 patients had either single-level LLIF, stand-alone LLIF or <4 levels of posterior instrumentation; 26 patients did not complete 2-year follow-up and follow-up imaging studies were not available for review in 28 patients and hence, these patients were excluded from the analysis) with complete 2-year follow-up. LLIF was performed at 637 levels. The mean patient age was 65.3 (SD 8.3) years; 74% of patients were females (46 males; 133 females). (Table 1)

Patient Characteristics

UIV was upper thoracic, the lower thoracic and upper lumbar vertebra in 25(14%), 118 (66%) and 36 (20%) patients respectively. LIV was L4-L6, sacrum and pelvis in 10 (5.5%), 81 (45.5%) and 88 (50%) patients respectively. The median length of instrumentation was 8 levels (IQR 7-9). Median interbody fusions were performed at 3 levels (IQR 3-4). The mean follow-up duration was 4.4 years (SD = 1.9). Osteotomies were employed in 38 (21%) patients; 30 (16.7%) patients had posterior column osteotomies (PCOs), and 8 (4.4%) patients had single-level pedicle subtraction osteotomy (PSO). Revision surgery was performed in 71 patients (39.6%); the cause of reoperation was proximal junctional kyphosis (PJK) in 16 (22.5%), proximal junctional fracture (PJF) in 21 (29.5%), rod fracture (other than LLIF-level) in 5 (7%), L5-S1 pseudoarthrosis in 18 (25.5%), thoracolumbar pseudoarthrosis in 3 (4.2%) and clinically significant pseudoarthrosis at LLIF-level in 2 (2.8%) patient.120 patients (67%) had postoperative CTs available for review during follow-up (Table 1). Of the 18 (25.5%) patients who were reoperated for L5-S1 pseudoarthrosis, 14 (78%) were without sacropelvic fixation and 17 (95%) had TLIF procedure at the L5-S1 level.

Group Characteristics

Baseline characteristics were similar in both fused and pseudoarthrosis groups. Three of the ten patients (33%) were obese (BMI>30) and 4/10 (40%) were smokers in the pseudoarthrosis group (Table 4). Both groups were similar in the number of interbody fusion levels (median = 3; IQR 3-4) and length of instrumentation (median = 8; IQR 7-9). All patients treated with osteotomies (n = 38) were in the fused group. Mean follow-up duration was not significantly different between the two groups (fused=4.4 years, pseudoarthrosis = 5 years; P = .266).

169 patients (94.5%) had radiological evidence of successful arthrodesis, while 10 patients (5.5%) had radiological pseudoarthrosis at one level of interbody fusion. Overall, pseudoarthroses were noted at 10 of 637 (.015%) operated levels. Given the available imaging modalities, pseudoarthrosis was determined by CT findings in 90% of patients in the pseudoarthrosis group, while 66% of patients in the fused group were regarded as ‘fused’ based on CT scans (Table 2).

Table 2.

Comparison of the Variables Between Fused and Pseudoarthrosis Group.

Variables		Radiological Arthrodesis (n = 179)		P-Value
		Present	Absent
Total, n (%)		169 (94.5)	10 (5.5)
Age (mean, SD)		65.1 (8.3)	69 (9.3)	.161
Sex, n (%)
	Males	43 (25.4)	3 (30)	.73
	Females	126 (74.6)	7 (70)
Upper instrumented vertebra (UIV), n (%)				1.0
	T1-T9	24 (14.2)	1 (9.0)
	T10-T12	111 (65.6)	7 (70)
	L1-L3	34 (20.1)	2 (20)
Lower instrumented vertebra (LIV), n (%)				.094
	L4-L6	8 (4.7)	2 (20)
	S1	76 (45)	5 (50)
	Ilium	85 (50.3)	3 (30)
Levels of instrumentation (median, IQR)		8 (7-9)	8 (6-9)	.731
Osteotomies, n(%)				<.001
	PCO	30	0
	PSO	8	0
Interbody fusion levels (median, IQR)		3 (3-4)	3 (3-4)	.924
Follow-up duration (years) (mean, SD)		4.4 (1.9)	5 (1.95)	.266
Follow-up CT, n(%)		111 (65.6)	9 (90)	.169

Abbreviations. PCO Posterior column osteotomy, PSO: pedicle subtraction osteotomy.

Of the 10 pseudarthroses patients, 8 (4.4%) were either clinically asymptomatic or had manageable back pain with complex medical issues at the last follow-up (Table 3). Two patients (1.1%) required revision surgery for symptomatic pseudoarthrosis at the LLIF-level. Both patients were reoperated with implant exchange and revision posterolateral fusion using rhBMP-2 (Figures 4-6). Pseudoarthrosis was not observed at more than one level of interbody fusion in the study population. Moreover, 5/10 (50%) patients demonstrated evidence of robust postero-lateral fusion.

Table 3.

Characteristics of Individual Patients with Pseudoarthrosis.

Id	Age	Sex	Prior Spine Surgery	Comorbidities		Smoking Status	Levels of Instrumentation		TLIF	LLIF	Levels of LLIF	ALIF	Revision Surgery	Cause of Reoperation					Follow-up Duration (Years)	Level of pseudoarthrosis	Radiological Findings of Pseudoarthroses	Time of Diagnosis since surgery (Years)	Clinical Status
Id	Age			Diabetes mellitus	Obesity		UIV	LIV						Pseudoarthrosis (other than LLIF Site)	PJK	PJF	Rod Fracture	Others (adjacent Segment Degeneration)
1	77	Male	No	No	No	No	T11	L5		T12-L5	5		No						6.4	L4-5	Lucency at L4 inferior end plate-spacer junction on CT enterography for UGIB	6.4	Asymptomatic
2	57	Female	Yes	No	No	Yes	T4	Ilium	L5-S1	L2-5	3		Yes	Yes	No	Yes	Yes	ASD	8.4	L2-3	Bilateral rod fractures on radiographs. Discontinuous and fragmented fusion mass in the intervertebral disc space observed on CT lumbar spine	5.1	Rod revision and postero-lateral fusion with rhBMP-2, inadequate fusion across disc space noted on last follow-up CT
3	50	Female	No	No	No	Yes	L2	S1	L5-S1	L2-5	3		Yes	No	No	Yes	No	Persistent sagittal imbalance	8.0	L2-3	Inadequate bridging bone trabeculations across the disc space	5.1	Required re-operations for PJF and persistent sagittal imbalance, died due to aspiration pneumonitis
4	75	Female	No	No	No	Yes	T10	S1	L5-S1	L2-5	3		Yes	Yes	No	Yes	Yes	No	5.0	L2-3	Inadequate bridging bone trabeculations across the disc space	4.9	Asymptomatic
5	71	Female	No	Yes	Yes	No	T11	L5		L2-5	3		No						3.1	L4-5	Inadequate bridging bone trabeculations across the disc space	2.2	Asymptomatic
6	72	Female	No	No	No	No	T10	S1	L5-S1	L1-5	4		No						3.3	L1-2	Right rod fracture	3.3	Right radiculopathy; L4-5 spacer subsidence with severe foraminal narrowing, no back pain
7	72	Male	No	Yes	Yes	No	L2	S1	L5-S1	L2-5	3		No						5.2	L2-3	Inadequate bridging bone trabeculations across the disc space, halo around left L2 screw	2.8	Complex medical issues, manageable back pain
8	72	Male	No	No	Yes	Yes	T10	Ilium		L1-5	4	L5-S1	Yes	Yes	No	Yes	No	No	4.3	L4-5, rod fracture at L5-S1	Inadequate bridging bone trabeculations across the disc space, bilateral rod fracture with loosened set screws noted on CT abdomen	3.9	Developed metastatic colo-rectal carcinoma, manageable back pain
9	64	Female	No	No	No	No	T10	Ilium		L2-5	3	L5-S1	No						4.0	L2-3	Bilateral rod fractures on radiographs. Discontinuous and fragmented fusion mass in the intervertebral disc space observed on CT lumbar spine	3.3	Rod revision and posterolateral fusion, improved back pain, asymptomatic PJK
10	80	Female	Yes	No	No	No	T10	S1		L1-4	3		No						3.0	L3-4	Inadequate bridging bone trabeculations across the disc space seen on CT abdomen done for evaluation of Leukemia	2.4	Asymptomatic

Figure 4.

(Patient 9): 64 years female with degenerative lumbar scoliosis (Pre-op coronal cobb:29, LL: 29, SS: 29, PI-LL mismatch: 31, SVA:84, PT:34) underwent L2-5 LLIF, L5-S1 ALIF and T10-Pelvis posterior instrumentation. Post-operative images (right) show restoration of lumbar lordosis and good correction of the sagittal alignment. The patient has mild persistent coronal imbalance to the right.

Figure 5.

(Patient 9 continued): The patient presented with progressive mechanical back pain and follow-up radiographs demonstrated bilateral rod fractures (arrows)at the L2-3 level. (Post-op coronal cobb:9, LL: 54, SS: 35, PI-LL mismatch: 10, SVA:43, PT:32). Computed tomography images show lucency in the disc space and at the spacer-end plate junction (right). The patient underwent re-operation with rod exchange and postero-lateral fusion using rhBMP-2.

Figure 6.

(Patient 9 continued): post-operative images obtained at 1-year follow-up after reoperation. The patient had excellent improvement in back pain symptoms. However, she persists to have asymptomatic proximal junctional kyphosis.

Radiographic Alignment Parameters

All patients in the study had significant improvement in coronal and sagittal alignment measures. Both groups were similar in their pre-operative alignment metrics. The postoperative change in alignment parameters was statistically significant (Table 4) for both groups. However, significant difference in mean postoperative SVA (pseudoarthrosis-group = 69.1; fused-group = 39.6 mm, P = .033), PT (pseudoarthrosis-group 23.7; fused-group 17.4, P = .025) and PI-LL mismatch (pseudoarthrosis-group 11; fused-group 1.1, P = .023) was noted between the groups (Table 5). Other parameters (TK, LL, and SS) were similar between the two groups.

Table 4.

Comparison of Pre-Operative and Post-Operative Alignment Measures for the Overall Study.

Radiographic Parameters	Pre-operative	Post-operative	P-Value
Coronal cobbs, degrees (mean, SD)	29.4 (14.2)	14.9 (10.5)	<.001
TK (mean, SD)	33.1 (15.9)	44.4 (14.2)	<.001
LL (mean, SD)	35.6 (18.1)	50.8 (13.1)	<.001
PI (mean, SD)	50.3 (11.7)
SVA (mean, SD)	52.7 (51.5)	41.8 (44.6)	.014
PT (mean, SD)	22 (9.2)	17.6 (8.6)	<.001
SS (mean, SD)	30 (11)	36.5 (10.3)	<.001
PI-LL mismatch (mean, SD)	15.4 (17.4)	1.7 (13.5)	<.001

TK: Thoracic kyphosis; LL: Lumbar lordosis; PI: Pelvic incidence; PT: Pelvic tilt; SS: Sacral slope (degrees); SVA: Sagittal vertical axis (mm).

Table 5.

Comparison of Pre-Operative and Post-Operative Measurement Between Fused and Pseudoarthrosis Group.

Radiographic Parameters	Pre-operative			Post-operative
	Fused Group	Pseudoarthrosis Group	P- Value	Fused group	Pseudoarthrosis Group	P- Value
Coronal cobbs (mean, SD)	29.6 (14.2)	24.7 (13.4)	.311	14.7 (10.3)	17.2 (13.3)	.467
TK (mean, SD)	33.2 (14.8)	31.4 (24.8)	.738	44.5 (14.1)	42.6 (13.8)	.691
LL (mean, SD)	35.7 (17.4)	32.3 (29)	.574	51 (13.3)	50.2 (10.5)	.866
PI (mean, SD)	50.6 (11.4)	48.4 (15.9)	.559
SVA (mean, SD)	51.3 (52.4)	65.23 (42.7)	.436	39.6 (42.3)	69.1 (56.2)	.033
PT (mean, SD)	22 (9.18)	21.5 (9.6)	.867	17.4 (8.4)	23.7 (9.8)	.025
SS (mean, SD)	29.8 (11)	30.9 (15.2)	.794	36.3 (10.3)	40.2 (11.6)	.245
PI-LL mismatch (mean, SD)	15.3 (17.3)	16.1 (19.8)	.886	1.1 (13)	11 (17.7)	.023

TK: Thoracic kyphosis; LL: Lumbar lordosis; PI: Pelvic incidence; PT: Pelvic tilt; SS: Sacral slope (degrees); SVA: Sagittal vertical axis (mm).

Discussion

Results from our study found a 94.5% arthrodesis rate for lateral lumbar interbody fusion (T12-L5) in the ASD cohort treated with multiple segment instrumentation. While 10 (5.5%) patients had radiological evidence of pseudoarthrosis, only two (1.1%) patients were significantly symptomatic requiring reoperation. Patients in the fused group achieved superior correction of SVA and PT and had lower PI-LL mismatch. Our results demonstrate that arthrodesis rates with multiple-level LLIF and rhBMP-2 in ASD are comparable to the previous studies of single-level LLIF. Retrospective reviews from prior studies have reported a 94%-96% arthrodesis rate with single level LLIF.^23,15

Several authors have compared lumbar arthrodesis rates between rhBMP-2 and ICBG in prior studies.^24,25 A recently published randomized control trial investigated the effect of rhBMP-2 in single-level PLIF procedure.²⁶ Forty patients were randomized in each group (rhBMP-2 and autograft) and patients achieved 100% arthrodesis in both the groups at 1-year follow-up. However, the authors noted a slower rate of trabecular bone formation and lower interbody fusion-mass density in the BMP group during the first year of surgery.²⁶ Multiple segment spinal fusion in ASD surgeries more frequently develops pseudoarthroses. Kim et al¹³ reported 17% (39/232) pseudarthrosis in their series of spinal deformity patients treated with multiple segment instrumentation and autologous bone-grafting.¹³ 50% of the pseudarthroses observed in their study occurred at the junctional region (seven at thoraco-lumbar, eighteen at the lumbosacral region). In contrast, we encountered 21 (11.7%) symptomatic pseudarthroses at junctional levels in our series which required re-operation. Given the higher reported rates of lumbosacral pseudoarthrosis in comparison to the other levels of the lumbar spine, L5-S1 fusions were omitted from the primary analysis.^27-30 Recently, Soroceanu et al³¹ described an 11.4% incidence of radiological pseudarthrosis in an ASD cohort, although the levels of pseudarthroses, the use of LLIFs and reoperation rates were not defined.³¹ How et al, in a recently published meta-analysis of 25 studies, reported a 6.3% combined revision rate for pseudoarthrosis in ASD patients.³² The radiological pseudoarthrosis at the interbody fusion level in our study was 5.5% and clinically significant pseudoarthrosis requiring re-operation was even lower (1.1%). The use of rhBMP-2 is associated with the high direct cost of the index procedure. However, rhBMP-2 is cost-effective in ASD surgeries in preventing pseudoarthrosis and re-operation.³³ The mean 2-year direct cost for patients with pseudarthrosis ($138,000$ ± 17 000) was significantly higher than that of patients who did not develop pseudarthrosis ($61,000 ± $25,000).³⁴

CT scans and radiographs are the most commonly employed imaging studies to assess fusion status.^35,36 Plain radiographs exhibit 89% sensitivity and 60% specificity for predicting solid arthrodesis in patients following PLFs.³⁷ Likewise, the sensitivity of CT scan to determine fusion status ranges from 70 to 90% and specificity ranges from 28 to 85% in patients with interbody fusion using metallic cages.³⁸ Peters et al, in a meta-analysis on assessment of fusion criteria,³⁹ demonstrated that CT-scan possessed the highest discriminant ability as against intra-operative exploration. Chun et al⁴⁰ performed a similar, yet a less-comprehensive review of imaging data and suggested that CT-scan has the highest sensitivity and specificity (53-63% and 86% respectively) among the imaging studies for the determination of arthrodesis.⁴⁰ While surgical exploration has historically been considered the ‘gold standard for the diagnosis of pseudoarthrosis¹⁹ in the PLF-era, it is not a practical alternative to confirm the diagnosis.

Due to the higher radiation exposure with CT scans, multiple studies for long-term follow-ups are not warranted on asymptomatic individuals. Radiographs are cheaper, easily accessible and pose lesser radiation exposure. We relied on a composite radiographic criterion derived from the literature to determine fusion status in our patients. Assessment of arthrodesis solely based on radiographic findings might have overestimated the arthrodesis rates in our cohort as 35% of patients in the fused group did not have a follow-up CT scan. However, these patients did not exhibit clinical or radiographic suspicion during follow-ups to merit CT imaging. Meanwhile, we had access to the lumbar spine CT images of several patients obtained for unrelated conditions. Imaging studies of each patient within the province could be accessed with a high capture rate using the unique patient identifier. We feel that such a captive population allows for high follow-up rates and a higher precision estimate of the true arthrodesis rate.

The mean time from index surgery to pseudoarthrosis detection in our study was 4.1 years (range 2.2-6.4) (Table 4). The mean follow-up duration for both study groups is over 4 years and the length of follow-up is reasonable to detect most of the pseudarthroses, although seven years of follow-up has been recommended in literature.⁴¹ Nevertheless, 8 patients (4.4%) with asymptomatic pseudoarthroses could become symptomatic in the longer term or these unfused segments with the potential fibrous union and radiologically demonstrable pseudoarthrosis could subsequently ossify into a stiff segment. A similar observation was made by Jung et al⁴¹ in their study of 152 patients treated with single or two-level LLIF which reported a pseudarthrosis rate of 27.6% (42/152) at 1-year follow-up, which decreased to 12.5% (19/152) at 2-year follow-up. Fourteen of these patients had asymptomatic, radiological pseudoarthrosis and 5 (3.2%) were symptomatic requiring re-operation. These authors also described multiple-level LLIFs, diabetes and smoking status as independent risk factors for pseudoarthrosis development.⁴¹ Five patients in the pseudoarthrosis group had posterolateral fusion despite the features of pseudoarthrosis at the interbody fusion level. This finding might explain the asymptomatic status of these patients and underlines the importance of meticulous PLF to supplement anterior interbody fusion.

Our strict definition of pseudoarthrosis identifiable at the interbody fusion levels could have resulted in a lower estimation of the pseudoarthrosis rate. We did observe five-rod fractures (2.42%) and junctional pseudoarthrosis but these rates are lower than that described in the literature.²⁷ Other limitations of the study were retrospective design, and lack of a control group to compare fusion rates with a fewer level of fusion or alternative biologics as described in the literature.^36,42 The study results were focused on the radiological findings and failed to describe clinical outcome scores after the procedure. Radiological and implant-related complications poorly affect the HRQOL outcome scores,³¹ and careful analyses of clinical outcomes would have helped to establish the impact of successful arthrodesis on quality of life. Lastly, the radiological criteria used in the study have not been validated to confirm pseudoarthrosis and CT images were available only for the two-thirds of the study cohort, which might have overestimated the true arthrodesis rate.

Despite these limitations, this study has notable strengths. It is the largest known series of ASD patients treated with multiple-level LLIFs. By limiting our definition to interbody fusion level, we have provided concise reporting of pseudoarthrosis at the lateral lumbar interbody fusion levels between the L1-L5 motion segment. Patients were followed prospectively with high image capture rates. The higher arthrodesis rate in our study could be attributed to the better disc space preparation and larger fusion surface provided by the lateral approach and rhBMP-2. A prospective study utilizing CT scans for evaluation of arthrodesis across fusion levels and longer follow-up data would be required to provide a precise estimate of the lumbar pseudoarthrosis rate.

Conclusion

LLIF along with rhBMP-2 can achieve high fusion rates in multi-segmental ASD surgeries. The pseudoarthrosis rate requiring reoperation in our study cohort was about 1%. Patients with persistent radiological signs of pseudoarthrosis may remain asymptomatic for several years.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Ethical Approval

IRB approval: REB19-0858, University of Calgary.

ORCID iDs

Vishwajeet Singh

Nathan Evaniew

References

Liu

Wang

Liang

Zhou

Chen

. Comparative clinical effectiveness and safety of bone morphogenetic protein versus autologous iliac crest bone graft in lumbar fusion: A meta-analysis and systematic review. Spine. 2020;45(12):E729-E741. doi:10.1097/BRS.0000000000003372.

Fernyhough

Schimandle

Weigel

Edwards

Levine

. Chronic donor site pain complicating bone graft harvesting from the posterior iliac crest for spinal fusion. Spine. 1992;17(12):1474-1480. doi:10.1097/00007632-199212000-00006.

Hoffmann

Jones

Sietsema

. Recombinant human bone morphogenetic protein-2 in posterolateral spinal fusion: What’s the right dose? Asian Spine J. 2016;10(3):457-464. doi:10.4184/asj.2016.10.3.457.

Adams

Ogden

Robertson

Broadhurst

Edis

. Effectiveness and safety of recombinant human bone morphogenetic protein-2 versus local bone graft in primary lumbar interbody fusions. Spine. 2014;39(2):164-171. doi:10.1097/BRS.0000000000000089.

Hurlbert

Alexander

Bailey

, et al. rhbmp-2 for posterolateral instrumented lumbar fusion: A multicenter prospective randomized controlled trial. Spine. 2013;38(25):2139-2148. doi:10.1097/BRS.0000000000000007.

Roh

Yeung

Field

McClellan

. Allogeneic morphogenetic protein vs. recombinant human bone morphogenetic protein-2 in lumbar interbody fusion procedures: A radiographic and economic analysis. J Orthop Surg. 2013;8:49. doi:10.1186/1749-799X-8-49.

Coughlan

Davies

Mostert

, et al. A prospective, randomized, multicenter study comparing silicated calcium phosphate versus BMP-2 synthetic bone graft in posterolateral instrumented lumbar fusion for degenerative spinal disorders. Spine. 2018;43(15):E860-E868. doi:10.1097/BRS.0000000000002678.

Galimberti

Lubelski

Healy

, et al. A systematic review of lumbar fusion rates with and without the use of rhBMP-2. Spine. 2015;40(14):1132-1139. doi:10.1097/BRS.0000000000000971.

Lee

Kim

. A comparison of radiographic and clinical outcomes of anterior lumbar interbody fusion performed with either a cellular bone allograft containing multipotent adult progenitor cells or recombinant human bone morphogenetic protein-2. J Orthop Surg. 2017;12(1):126. doi:10.1186/s13018-017-0618-8.

10.

Hsu

Wang

. The use of bone morphogenetic protein in spine fusion. Spine J. 2008;8(3):419-425. doi:10.1016/j.spinee.2008.01.008.

11.

Emami

Deviren

Berven

Smith

Bradford

. Outcome and complications of long fusions to the sacrum in adult spine deformity: Luque-galveston, combined iliac and sacral screws, and sacral fixation. Spine. 2002;27(7):776-786. doi:10.1097/00007632-200204010-00017.

12.

Takahashi

Delécrin

Passuti

. Surgical treatment of idiopathic scoliosis in adults: An age-related analysis of outcome. Spine. 2002;27(16):1742-1748. doi:10.1097/00007632-200208150-00011.

13.

Kim

Bridwell

Lenke

Cho

Edwards

Rinella

. Pseudarthrosis in adult spinal deformity following multisegmental instrumentation and arthrodesis. J Bone Joint Surg Am. 2006;88(4):721-728. doi:10.2106/JBJS.E.00550.

14.

Maeda

Buchowski

Kim

Mishiro

Bridwell

. Long adult spinal deformity fusion to the sacrum using rhBMP-2 versus autogenous iliac crest bone graft. Spine. 2009;34(20):2205-2212. doi:10.1097/BRS.0b013e3181b0485c.

15.

Nourian

Harrington

Pulido

McCauley

Bruffey

Eastlack

. Fusion rates of lateral lumbar interbody fusion using recombinant human bone morphogenetic protein-2. Global Spine J. 2019;9(4):398-402. doi:10.1177/2192568218797097.

16.

Taba

Williams

. Lateral lumbar interbody fusion. Neurosurg Clin N Am. 2020;31(1):33-42. doi:10.1016/j.nec.2019.08.004.

17.

Lee

Kang

, et al. Minimally invasive multilevel lateral lumbar interbody fusion with posterior column osteotomy compared with pedicle subtraction osteotomy for adult spinal deformity. Spine J. 2020;20(6):925-933. doi:10.1016/j.spinee.2019.12.001.

18.

Hilibrand

Dina

. The use of diagnostic imaging to assess spinal arthrodesis. Orthop Clin North Am. 1998;29(4):591-601. doi:10.1016/s0030-5898(05)70033-x.

19.

Burkus

Foley

Haid

LeHuec

. Surgical interbody research group--radiographic assessment of interbody fusion devices: fusion criteria for anterior lumbar interbody surgery. Neurosurg Focus. 2001;10(4):E11-E19. doi:10.3171/foc.2001.10.4.12.

20.

Pai

Hodgson

. Assessment of bony union following surgical stabilisation for lumbar spondylolysis: a comparative study between radiography and computed tomography. J Orthop Surg. 2006;14(1):17-20. doi:10.1177/230949900601400105.

21.

Williams

Gornet

Burkus

. CT evaluation of lumbar interbody fusion: Current concepts. AJNR Am J Neuroradiol. 2005;26(8):2057-2066.

22.

Malham

Parker

Blecher

Seex

. Assessment and classification of subsidence after lateral interbody fusion using serial computed tomography. J Neurosurg Spine. 2015;23(5):589-597. doi:10.3171/2015.1.SPINE14566.

23.

Kiely

Bogner

, et al. Early clinical and radiological results of unilateral posterior pedicle instrumentation through a Wiltse approach with lateral lumbar interbody fusion. J Spine Surg. 2017;3(3):338-348. doi:10.21037/jss.2017.06.16.

24.

Khan

Pearce

McAnany

Peters

Gupta

Zebala

. Comparison of transforaminal lumbar interbody fusion outcomes in patients receiving rhBMP-2 versus autograft. Spine J. 2018;18(3):439-446. doi:10.1016/j.spinee.2017.08.230.

25.

Adams

Ogden

Robertson

Broadhurst

Edis

. Effectiveness and safety of recombinant human bone morphogenetic protein-2 versus local bone graft in primary lumbar interbody fusions. Spine. 2014;39(2):164-171. doi:10.1097/BRS.0000000000000089.

26.

Michielsen

Sys

Rigaux

Bertrand

. The effect of recombinant human bone morphogenetic protein-2 in single-level posterior lumbar interbody arthrodesis. J Bone Joint Surg Am. 2013;95(10):873-880. doi:10.2106/JBJS.L.00137.

27.

Kim

Bridwell

Lenke

Rhim

Cheh

. Pseudarthrosis in long adult spinal deformity instrumentation and fusion to the sacrum: prevalence and risk factor analysis of 144 cases. Spine. 2006;31(20):2329-2336. doi:10.1097/01.brs.0000238968.82799.d9.

28.

Harimaya

Mishiro

Lenke

Bridwell

Koester

Sides

. Etiology and revision surgical strategies in failed lumbosacral fixation of adult spinal deformity constructs. Spine. 2011;36(20):1701-1710. doi:10.1097/BRS.0b013e3182257eaf.

29.

Annis

Brodke

Spiker

Daubs

Lawrence

. The Fate of L5-S1 with low-dose BMP-2 and pelvic fixation, with or without interbody fusion, in adult deformity surgery. Spine. 2015;40(11):E634-E639. doi:10.1097/BRS.0000000000000867.

30.

Lee

Kang

, et al. Strategy for obtaining solid fusion at L5-S1 in adult spinal deformity: Risk factor analysis for nonunion at L5-S1. J Neurosurg Spine. 2020;33:323-331. doi:10.3171/2020.2.SPINE191181.

31.

Soroceanu

Diebo

Burton

, et al. radiographical and implant-related complications in adult spinal deformity surgery: Incidence, patient risk factors, and impact on health-related quality of life. Spine. 2015;40(18):1414-1421. doi:10.1097/BRS.0000000000001020.

32.

How

Street

Dvorak

, et al. Pseudarthrosis in adult and pediatric spinal deformity surgery: A systematic review of the literature and meta-analysis of incidence, characteristics, and risk factors. Neurosurg Rev. 2019;42(2):319-336. doi:10.1007/s10143-018-0951-3.

33.

Safaee

Dalle Ore

Zygourakis

Deviren

Ames

. Estimating a price point for cost-benefit of bone morphogenetic protein in pseudarthrosis prevention for adult spinal deformity surgery. J Neurosurg Spine. Epub ahead of print 2019. doi:10.3171/2018.12.SPINE18613.

34.

Jain

Yeramaneni

Kebaish

, et al. Cost-utility analysis of rhBMP-2 use in adult spinal deformity surgery. Spine. 2020;45(14):1009-1015. doi:10.1097/BRS.0000000000003442.

35.

Rodgers

Gerber

Patterson

. Fusion after minimally disruptive anterior lumbar interbody fusion: Analysis of extreme lateral interbody fusion by computed tomography. SAS J. 2010;4(2):63-66. doi:10.1016/j.esas.2010.03.001.

36.

Pimenta

Marchi

Oliveira

Coutinho

Amaral

. A prospective, randomized, controlled trial comparing radiographic and clinical outcomes between stand-alone lateral interbody lumbar fusion with either silicate calcium phosphate or rh-BMP2. J Neurol Surg A Cent Eur Neurosurg. 2013;74(6):343-350. doi:10.1055/s-0032-1333420.

37.

Brodsky

Kovalsky

Khalil

. Correlation of radiologic assessment of lumbar spine fusions with surgical exploration. Spine. 1991;16(6 suppl l):S261-S265. doi:10.1097/00007632-199106001-00017.

38.

Carreon

Glassman

Schwender

Subach

Gornet

Ohno

. Reliability and accuracy of fine-cut computed tomography scans to determine the status of anterior interbody fusions with metallic cages. Spine J. 2008;8(6):998-1002. doi:10.1016/j.spinee.2007.12.004.

39.

Peters

MJM

Bastiaenen

CHG

Brans

Weijers

Willems

. The diagnostic accuracy of imaging modalities to detect pseudarthrosis after spinal fusion-a systematic review and meta-analysis of the literature. Skeletal Radiol. 2019;48(10):1499-1510. doi:10.1007/s00256-019-03181-5.

40.

Chun

Baker

Hsu

. Lumbar pseudarthrosis: A review of current diagnosis and treatment. Neurosurg Focus. 2015;39(4):E10. doi:10.3171/2015.7.FOCUS15292.

41.

Jung

Chung

Kim

Yang

. Prognosis of symptomatic pseudarthrosis observed at 1 year after lateral lumbar interbody fusion. Spine. 2021;46(18):E1006-E1013. doi:10.1097/BRS.0000000000003980.

42.

Parker

Malham

. Comparison of a calcium phosphate bone substitute with recombinant human bone morphogenetic protein-2: A prospective study of fusion rates, clinical outcomes and complications with 24-month follow-up. Eur Spine J. 2017;26(3):754-763. doi:10.1007/s00586-016-4927-0.