Preserving spinal motion: lumbar disc arthroplasty in the US

Abstract

Symptomatic lumbar degenerative disc disease affects millions of patients, is a leading cause of physician visits annually, and significantly impacts healthcare systems globally. The gold standard treatment for this condition is spinal fusion, but its mechanism of eliminating segmental motion and risk of adjacent segment disease resulting in additional surgery has increased interest in alternative treatments. Lumbar disc arthroplasty has emerged as an alternative to fusion due to its comparable patient outcomes coupled with the ability to preserve spinal motion resulting in greater long-term benefit potential. Yet, widespread adoption of the procedure has been dampened by low reimbursement rates and disproportionate insurance denials compared to lumbar fusions, affecting patient access to this technology. Artificial disc procedures have been performed in Europe since the mid-1980’s with favorable results, leaving the United States comparatively behind. The aim of this review article is to summarize the current literature regarding lumbar disc arthroplasty as it pertains to implants available in the United States.

INTRODUCTION

The prevalence of low back pain continues to be on the rise, affecting approximately 620 million people globally in 2020 and an estimated 840 million people by 2050.¹ The prevalence of lumbar degenerative disc disease (DDD) continues to increase and is responsible for millions of physician visits for episodic low back pain in the US annually.^2,3 Lumbar fusion is the gold standard treatment for refractory symptomatic lumbar DDD, ceasing motion and providing stability of the affected segment. In 2015, the average aggregate hospital cost was estimated to be over $50,000 per admission for lumbar fusion and over 400,000 lumbar fusions are performed annually.^4,5 While fusion has been effective at reducing symptoms related to DDD, the elimination of segmental motion significantly alters the biomechanics of the spine, which can have ramifications related to patient activities, quality of life, adjacent segment stress, the need for additional surgery, and increased healthcare costs.

Lumbar disc arthroplasty (LDA) is a rising alternative treatment for lumbar DDD that has gained recognition due to its disc space height restoration, comparable patient reported outcomes to fusion, and unique ability to preserve or improve spinal motion thereby minimizing adjacent segment disease (ASD) and additional surgery.⁶ Although fusion and arthroplasty achieve comparable short-term objectives of removing damaged discs and alleviating symptoms, their trajectories diverge in the long term, particularly with respect to reducing the progression of ASD and reducing the need for reoperation, as well as maintaining the health benefits associated with sustained physical activity.

After a brief history of LDA, this review article will provide an in-depth summary of surgical, clinical, radiographic, and economic outcomes from the earliest clinical trials to the more recent observational studies pertaining to the artificial disc implants available in the US today. By providing the state of the evidence, this review article aims to be a useful tool for all stakeholders.

BRIEF HISTORY OF LUMBAR DISC ARTHROPLASTY

The first recorded lumbar disc arthroplasty was performed by Ulf Fernström in 1960 by implanting stainless steel spheres into a lumbar disc space.⁷ Subsidence and extrusion occurred, limiting further research and application of this procedure. In 1982, Kurt Schellnack and Karin Buettner-Janz developed the SB Charité disc, a precursor to the artificial discs of today. The SB Charité design featured two stainless steel endplates and an interposed, sliding polyethylene core. Years of design refinement and positive preliminary results led to the development of new artificial discs, each differing in design, construct, and biomechanical characteristics.⁸ Although spinal arthroplasty procedures have been performed in Europe regularly since the 1990’s, the first lumbar artificial disc was not approved by the U.S. Food and Drug Administration (FDA) until 2004.

FDA APPROVED LUMBAR ARTIFICIAL DISCS

In 2004, the Charité® III (DePuy Spine, Raynham, MA, USA) became the first lumbar artificial disc approved by the FDA for single-level use at L4/5 or L5/S1 for the treatment of DDD.​​ Later rebranded as INMOTION®, this third generation design incorporated two cobalt-chrome alloy endplates and a free-floating ultra-high molecular weight polyethylene core. The mobile, sliding core promoted better flexion and extension of the lumbar spine. Despite these advancements, mixed clinical outcomes and complications led to the discontinuation of the Charité disc which was ultimately taken off the market in 2010.^9–11

In 2006, the ProDisc®L (DePuy Synthes Spine, West Chester, PA, USA) was approved by the FDA for single-level use at L3/4, L4/5, or L5/S1 for the treatment of DDD.​​ In 2020, ProDisc-L became the only lumbar artificial disc with FDA approval for the treatment of 2 consecutive levels (L3-S1).​ ProDisc-L features two cobalt chrome alloy endplates and an ultra-high molecular weight polyethylene core inlay, following a ball-and-socket design with a fixed center of rotation utilizing a spiked central keel.^9,12 An example of ProDisc-L at L4/5 can be seen in Figure 1.

FIGURE 1.

Standing anteroposterior (left) and lateral (right) radiographs depicting lumbar disc arthroplasty at L4/5 with ProDisc-L.

In 2015, the activL® (Aesculap Implant Systems, Center Valley, PA, USA) was approved by the FDA for single-level use at L4/5 or L5/S1 for the treatment of lumbar DDD.​​ The activL disc consists of two cobalt chromium alloy endplates with titanium and dicalcium phosphate coating, three anterior horizontal spikes or a central keel for internal fixation, and one ultra-high molecular weight polyethylene core. Unlike the fixed center of rotation of ProDisc-L, activL utilizes a partially restricted polyethylene core that supports translation in the anteroposterior direction, simulating physiological motion.^2,9 The option of selecting an implant with spikes or central keel for fixation allows the surgeon to customize the implant based on endplate morphology, which is particularly relevant for the superior endplate of S1. An example of activL at L5/S1 can be seen in Figure 2.

FIGURE 2.

Standing anteroposterior (left) and lateral (right) radiographs depicting lumbar disc arthroplasty at L5/S1 with activL.

The current FDA approvals with their indications and contraindications are summarized in Table 1. Although not the focus of this review, it is important to mention that other lumbar artificial discs were not successful at achieving FDA approval and new devices are currently undergoing clinical trials in the US to obtain FDA approval as the market continues to expand.

TABLE 1.

Current FDA approved artificial discs in the United States

Disc	FDA approval year	Indications	Contraindications	Approved levels	Design description
ProDisc-L	2006 (1-level) 2020 (2-levels)	Lumbar DDD at 1-2 contiguous level(s) < Grade 1 spondylolisthesis at affected level(s) Failed ≥ 6 mo conservative treatment	T-score < -1.0 Active infection or allergies to implant materials Bony lumbar stenosis or isolated radicular compression syndromes Degenerative or lytic spondylolisthesis of grade >1 Vertebral endplate is <34.5 mm in medial-lateral and/or 27 mm in the anterior-posterior directions	L3/4  L4/5  L5/S1	2 CCA endplates and UHMWPC inlay, following a ball and socket design with a fixed center of rotation
activL	2015 (1- level)	Lumbar DDD at 1-level <Grade 1 spondylolisthesis at affected level(s) Failed ≥ 6 mo conservative treatment	T-score ≤ -1.0 Active infection or allergies to implant materials Isolated lumbar radiculopathy, chronic radiculopathy (≥1 yr) Myelopathy or spinal stenosis Displaced disc material with sequestrum	L4/5  L5/S1	2 CCA endplates with titanium and dicalcium phosphate, 3 anterior horizontal spikes or central keel for internal fixation, and 1 UHMWPC core. Utilizes a partially restricted polyethylene core

Abbreviation: CCA, cobalt chrome alloy; DDD, degenerative disc disease; FDA, Food and Drug Administration; UHMWPC, ultra-high molecular weight polyethylene core.

FDA IDE TRIALS

ProDisc-L

In 2003, early results were reported in an interim analysis of the FDA Investigational Device Exemption (IDE) randomized controlled trial (RCT) for ProDisc-L. This study compared outcomes of circumferential fusions with LDA up to 6 mo on the first 53 patients. Surgically, there were no instances of implant migration, failure, or revisions. Clinically, VAS and ODI improved faster in the LDA group, but significant differences between groups were lost by 6 mo. Radiographically, significant improvement in range of motion (ROM) was seen at L4/5 in the LDA group (P<0.04). Overall, this interim analysis suggested noninferiority of LDA, and the trial continued.¹³

In 2007, Zigler et al. published the final results of the ProDisc-L IDE RCT comparing single level LDA (n=211) to circumferential fusion (n=75) up to 2 yr. Surgically, there were no major complications, but LDA had significantly lower blood loss, operative times, and hospital lengths of stay (LOS) compared to fusion (P<0.001 for all comparisons). The index level reoperation rate was 3.7% for LDA and 5.4% for fusion. Clinically, ODI scores were significantly improved from baseline in both groups up to 2 yr (P<0.001), but LDA outperformed fusion in achieving the minimum clinically important difference (MCID, P=0.045). VAS pain scores significantly improved from baseline in both groups, but patient satisfaction favored LDA at 2 yr (P=0.015). Short Form (SF-36) scores significantly improved early for LDA, which remained higher but lost significance at 2 yr. LDA outperformed fusion in the predefined overall success category up to 2 yr (P=0.044) and for patients confirming at 1 yr that they would have surgery again (P<0.001). At 2 yr, 92.4% of LDA patients were employed compared to 85.1% of fusion patients (P=0.049) and had better recreation status (P=0.031). Radiographically, index level ROM was restored to the predefined normal motion in 93.7% of LDA patients, and 97% of fusion patients met the fusion criteria.¹⁴

In 2007, 3-year results of the ProDisc-L IDE RCT included additional patients enrolled in the continued-access phase of the trial to compare LDA (n=126) to circumferential fusion (n=30). Surgically, LDA had significantly less blood loss, operative time, and hospital LOS compared to fusion (P<0.01 for all comparisons). Clinically, though LDA consistently scored better than fusion for VAS and ODI scores, both groups saw significant improvements at 3 yr compared to baseline. Both groups scored high on patient satisfaction, though only significantly higher at 2 yr in the LDA group, and responses were significantly more favorable at all time points (P<0.05) for LDA when asked about having the procedure again.¹⁵

In 2012, a radiographic post hoc analysis of 5-year data from the ProDisc-L IDE RCT was reported for single-level cases (123 LDA, 43 fusion cases). Surgically, adjacent-level surgery occurred less in the LDA group (1.6% LDA, 4.0% fusion, P=0.68). Radiographically, the presence of ASD was not different between groups prior to surgery (P=0.14), but there was a significant difference between the presence of ASD at 5 yr between groups (change in ASD observed: 9.2% LDA, 28.6% fusion, P<0.01). In fact, fusion patients were found to have 4.5 times greater likelihood of worsening ASD compared to those in the LDA group. At 5 yr, index level motion was decreased in both groups (-1.3° LDA, -6.0° fusion) compared to baseline motion. At the superior adjacent segment, motion somewhat increased in both groups (+0.3° LDA, +1.8° fusion).¹⁶

ActivL

In 2015, Garcia et al. published 2-year outcomes of the FDA IDE RCT for activL. This study compared outcomes of 1-level LDA procedures between activL (n=218) and a control group (ProDisc-L or Charité, n=106). Noninferiority was assessed at the interim analysis, then superiority was assessed at the final analysis for the primary outcome of treatment success defined as ≥ 15-point ODI improvement, no neurologic decline, no decrease in index level ROM, no additional index level surgery, and no device-related serious adverse events. Clinically, a higher treatment success rate was reported with activL compared to controls (P=0.02) at 2 yr. VAS back pain, ODI, quality of life scores, patient satisfaction, return to work, and narcotic consumption were assessed at 2 yr and improvements were not significantly different between LDA groups. Radiographically, change in index level ROM favored the activL group (+0.9° activL, -1.4° controls, P<0.01), but heterotopic ossification was similar between groups (1.6% activL, 1.1% controls). Surgically, the rates of revision, reoperation, removal, or supplemental fixation was similar between groups (4.4% activL, 3.5% control) at 2 yr.¹⁷

In 2019, Yue et al. published the 5-year outcomes of the activL IDE RCT comparing activL (n=218) to LDA controls (n=106). Clinically, both LDA treatment groups reached the primary outcome of treatment success, both groups saw a significant improvement in VAS back pain and ODI from baseline, and <2% of patients in both groups were taking narcotics at 5 yr compared to >60% preoperatively. Radiographically, the activL group showed significantly greater ROM at 5 yr compared to the LDA controls (P=0.02). Surgically, device-related serious adverse events and index level reoperation were low in both groups. Adjacent level additional surgery was 1% activL and 6% in the control group (P=0.01).¹⁸

In 2021, 7-year results from the FDA IDE RCT were compared between activL (n=218) and ProDisc-L (n=65), which excluded the Charité controls because it was no longer commercially available. Clinically, both LDA treatment groups reached the primary outcome of treatment success. Radiographically, the median ROM at baseline was similar between groups and remained similar between groups at 7 yr (P=0.18). Evidence of heterotopic ossification was not significantly different between groups at 7 yr. Surgically, the rate of reoperation was very low in both groups (5% activL, 0% ProDisc-L, P>0.05) at 7 yr.¹⁹

PROSPECTIVE OBSERVATIONAL STUDIES

While the IDE trials demonstrated safety and efficacy of LDA using ProDisc-L and activL to 7 yr postoperatively, known limitations of IDE trials exist. Namely, the strict inclusion criteria often do not reflect real-world practice, and the question of generalizability is often raised. Several prospective observational studies involving ProDisc-L and activL with expanded inclusion criteria and indications have been conducted.

ProDisc-L

A large prospective study assessed LDA cases (n=181) up to 10 yr. Treated levels included L2/3, L3/4, L4/5, L5/6, and L6/S1, and the proportion of levels treated was 83.4% 1-level, 16% 2-levels, and 0.6% 3-levels. Indications expanded from DDD alone to also include herniation and previous discectomy. Clinically, improvements in ODI and VAS were significant compared to baseline up to 10 yr (P<0.001). Patient satisfaction at 10 yr was 86.3%, those willing to undergo surgery again was 79.3%, and the rate of patients resuming professional activity was 66.9%. Surgically, the overall complication rate was 14.4% and the reoperation rate was 16%. Surgeries performed for adjacent level pathologies (2.2%) occurred beyond 6 yr after the index surgery.²⁰

Several years later, another large prospective cohort study of LDA patients (n=159) was conducted, having 2-4 consecutive levels treated and 2 yr of follow-up. The proportion of levels treated was 71.7% 2-levels, 25.8% 3-levels, and 2.5% 4-levels. Outcomes between these groups were compared. Clinically, all groups exhibited significant improvements in ODI, VAS pain, and satisfaction at 2 yr compared to baseline. Radiographically, adjacent level ROM did not significantly increase (+0.2°) compared to baseline in any group. Excluding L5/S1, all treated levels saw a significant increase in ROM (+2.2°, P<0.05 for all comparisons) compared to baseline. At L5/S1, there was a decrease in ROM (-1.8°, P=0.60) compared to baseline. Surgically, 1.9% underwent reoperation of the index level, 0% had adjacent level surgery, and 4.4% had reported complications up to 2 yr.²¹

ActivL

In 2015, Lu et al. conducted a small prospective cohort study of LDA patients (n= 30) with 3 yr of follow-up. Treated levels included L3/4, L4/5, and L5/S1 and the proportion of levels treated was 80% 1-level and 20% 2-levels. Clinically, treatment success (defined by IDE trial criteria) was reported in 86.7% of patients at 3 yr. Compared to baseline, significant improvements in VAS back and leg pain and ODI scores were maintained to 3 yr (P<0.001 for all comparisons). All patients were employed full-time preoperatively, and only 13.3% were not working or had no plans of working at final follow up. Surgically, no device failures, major complications, or revision surgeries occurred at 3 yr. Radiographically, asymptomatic implant subsidence and heterotopic ossification were separately observed in 10% and 3.3% of patients, respectively. Index level ROM was significantly increased (+2.2°, P=0.013) compared to baseline, but adjacent level ROM also increased (+3.4°, P<0.001) compared to baseline.²²

In 2020, another small prospective cohort study assessed outcomes of single level LDA patients (n=33) with a mean follow up of 2.7 yr. Clinically, ODI scores significantly improved from baseline at each milestone until the final follow up (P<0.001). There was a significant reduction in patients reporting back and leg pain at final follow up (P<0.001 for both comparisons). Preoperatively, 33% of patients were unable to work due to their back condition and 100% were employed at final follow up. Opioid use significantly decreased from baseline to final follow up (P<0.001). Surgically, no index level reoperations occurred to final follow up. Radiographically, only 1 patient developed mild subsidence that did not require surgical intervention, while 0% of the remaining patients developed implant migration or motion restriction up to final follow up.²³

RETROSPECTIVE STUDIES

Although retrospective studies are associated with lower levels of evidence due to several inherent study design limitations, the studies presented here include those of long-term follow up or present new outcomes, such as spinal alignment and cost analysis. These additional findings are useful to consider.

ProDisc-L

A retrospective radiographic analysis evaluated 58 levels of LDA (n=42) up to 11 yr. Treated levels included L2/3, L3/4, L4/5, and L5/S1 and the proportion of levels treated was 64.3% 1-level, 28.6% 2-levels, and 7.1% 3-levels. Mean follow-up was 8.7 yr. Radiographically, 24% developed signs of ASD. Mean index level ROM averaged 1.6° in those that developed ASD and 4.7° in those that did not develop ASD. In fact, no patients with index level ROM ≥ 5° developed ASD at 7 yr, and the odds of developing ASD if index level ROM ≤ 5° was 13.5 times higher (P=0.021). Clinically, no differences were reported between groups that developed ASD compared to those that did not up to 8.7 yr.²⁴

Spinal alignment was retrospectively reviewed for LDA cases (n=26) at 3 yr. Treated levels included L3/4, L4/5, and L5/S1 and the proportion of levels treated was 73.1% 1-level and 26.9% 2-levels. Radiographically, for single level LDA at L4/5 compared to baseline, there was a significant increase in segmental and lumbar lordosis (P<0.001 and P=0.006, respectively), but there was no difference in other alignment parameters. For single level LDA at L5/S1 compared to baseline, there was a significant increase in segmental lordosis at L5/S1 (P=0.017), but no difference in other parameters. Index level ROM significantly increased from baseline at both L4/5 and L5/S1 (P=0.008 and P=0.006, respectively) but was not reported for L3/4.²⁵

A small retrospective cost analysis of single center data from the ProDisc-L IDE RCT compared hospital charges between LDA (n=35) and circumferential fusion (n=17) groups. Economically (1-level cases), significant differences were found between groups in all charge categories, with a mean total charge for LDA at $35,592 compared to fusion at $46,280 (P<0.002), representing a 23% decrease in total charges for LDA. Surgeon fees were significantly lower in the LDA group ($1413 LDA, $4917 fusion, P<0.001). Economically (2-level cases), total charges were similar between groups, with a 2% decrease in total charges for LDA, however surgeon fees remained significantly lower for LDA ($2826 LDA, $5857 fusion, P<0.001).²⁶

In 2010, Buttacavoli et al. published a retrospective cost comparison between 3-level LDA (n=21) and 3-level circumferential fusion (n=22). Baseline characteristics differed by age (P<0.01) and worker’s compensation coverage (P<0.01) between groups but were otherwise balanced. Surgically, operating time was 202.9 minutes longer in the fusion group, blood loss was 1,634 mL greater in the fusion group, and hospital LOS was 3.2 days longer in the fusion group (P<0.001 for all comparisons). Economically, the cost of 3-level fusion was more than twice the cost of 3-level LDA ($81,499.05 LDA, $174,010.35 fusion, P<0.05). Fusion implant costs were also more than twice that of LDA ($27,000.00 LDA, $70,460.00 fusion, P<0.001).²⁷

ActivL

A retrospective review included cases collected from a single center having either LDA (n=30) or ALIF (n=50) for single level disease at L4/5 or L5/S1 with 1 yr follow-up. Clinically, there was no significant difference in baseline ODI and VAS back pain scores between groups, yet LDA had significantly improved VAS back pain and ODI scores compared to fusion up to 1 yr (P=0.007 and P=0.001, respectively). LDA patients returned to work 65 days sooner than fusion patients (P=0.011).²⁸

To evaluate ASD, Zigler et al. published a post hoc analysis of single level LDA cases from the activL IDE RCT, (activL, n=136 and ProDisc-L, n=39) with 5-year follow-up, which also included a matched comparison fusion group. Radiographically, 90.3% of all LDA patients showed no change in ASD at the superior adjacent level at 5 yr. Increased index level ROM ranged from 0° to 16.1° at 5 yr, and for each additional degree of ROM gained at the LDA level, there was a consistent decrease in percentage of change in ASD. Compared to the matched fusion group, LDA had a significantly lower likelihood of a change in ASD (OR 0.32; 95% CI 0.13-0.76). Surgically, 2.3% of LDA patients had adjacent level surgery. When compared to the matched fusion group, the likelihood of adjacent level surgery was lower in the LDA group, but this did not achieve significance (OR 0.76; 95% CI 0.42-1.36).²⁹

Artificial Disc Unspecified

In a retrospective radiographic review to evaluate ASD, 1000 LDA cases and 67 fusions were compared. Surgically, 2% in the LDA group underwent additional surgery for ASD over 2 yr after the index operation, and 4.5% of the fusion cases underwent additional surgery for ASD at about 5 yr after the index operation.³⁰

A recent retrospective database study compared propensity matched 1- and 2-level LDA (n=1,625) with stand-alone anterior lumbar interbody fusion (ALIF, n=1,625) cases to assess the rate of ASD at 3 yr. In this database study, ASD was defined as any subsequent arthrodesis, decompression, or spinal instrumentation surgery at the adjacent level within 3 yr of index surgery. Surgically, there were no differences in complications between groups, but ALIF was associated with longer hospital LOS (P<0.001). The rate of ASD surgery was over 2 times higher in the ALIF group compared to LDA at 3 yr (6.6% LDA, 14.5% ALIF, P<0.001) and the average time to ASD surgery was 3.9 yr in the ALIF group and 9.2 yr in the LDA group. Economically, LDA was associated with significantly reduced direct hospital costs compared to fusion ($49,000 LDA, $195,000 ALIF, P<0.001).³¹

In 2024, Guyer et al. published a large retrospective study of LDA cases (n=2,141) from a multi-surgeon center to investigate the incidence of index level reoperation up to 20 yr. Surgically, 1.3% of patients underwent index level reoperation at final follow-up. Of the 2,513 devices implanted in this cohort, 25 (0.99%) devices were removed and 3 (0.12%) were revised. Of those having index level reoperation, 37% occurred within one month after index implantation, the median time to reoperation was 9.4 mo, but only 1 reoperation occurred in the long-term (at 15.4 yr).³²

HYBRID RESULTS

In addition to studies with expanded indications and inclusion criteria compared to the narrow IDE trials, hybrid surgery is another off-label use of LDA that has become increasingly popular. Hybrid surgery combines at least one level of arthroplasty with at least one level of fusion, giving surgeons the ability to further customize spine surgery according to specific patient needs per level. An example of a 2-level hybrid surgery can be seen in Figure 3 and an example of 3-level hybrid surgery can be seen in Figure 4.

FIGURE 3.

Standing anteroposterior (left) and lateral (right) radiographs depicting 2-level hybrid surgery of lumbar disc arthroplasty at L4/5 with ProDisc-L and anterior lumbar interbody fusion with plate at L5/S1.

FIGURE 4.

Standing anteroposterior (left) and lateral (right) radiographs depicting 3-level hybrid surgery of lumbar disc arthroplasty at L3/4 and L4/5 with ProDisc-L and anterior lumbar interbody fusion with plate at L5/S1.

In 2021, Cuellar et al. published a prospective cohort study of multilevel hybrid surgery cases (n=46) with up to 6 yr of follow-up. All cases included an ALIF at L5/S1 combined with ProDisc-L at either L4/5 (n=23), L3/4 and L4/5 (n=19), or L2/3, L3/4, and L4/5 (n=4). Radiographically, ROM at all arthroplasty levels significantly increased compared to baseline (P=0.011), which was similar between hybrid groups. All fusion levels had motion eliminated (solid fusion). At all superior adjacent levels, there was no significant difference in ROM compared to baseline (P=0.21). Clinically, all hybrid groups achieved significant improvements in ODI and VAS pain scores compared to baseline, but no difference was found between groups. Satisfaction at final follow up did not achieve significance in any hybrid group, but 100% of patients responded positively to choosing the surgery again. Surgically, 0% of patients had major complications, reoperation, or device failures to final follow-up.³³

In 2022, Scott-Young et al. published a prospective cohort study comparing outcomes between single level LDA (n=211), multilevel LDA (n=122), and hybrid constructs (combination of LDA and stand-alone ALIF, n=617) with follow-up ranging from 2 to 10 yr. Surgically, there were no revisions or reoperations of fused levels and very low rates of revision or reoperation between arthroplasty groups (single level 2.8%, multilevel 5.4%, and hybrid 3.7%). Index level revisions occurred at a median time of nearly 3 yr. Clinically, significant improvements from baseline for VAS back pain, VAS leg pain, ODI, and satisfaction were found in all groups, and all were above the MCID for each applicable measure.³⁴

In 2024, Shellack et al. published a retrospective study reporting long-term outcomes between two groups (follow-up ≥ 10 yr, n=51 and follow-up < 10 yr, n=254) for hybrid surgery patients. In this study, all hybrid constructs consisted of one arthroplasty level and 1- or 2-level ALIF (standalone or circumferential fusion). Clinically, both groups had significant improvement in back pain, leg pain, and ODI scores compared to baseline (P<0.01 for all comparisons) and no differences were found between groups. Surgically, reoperation occurred in 16% of patients overall, of which 32.7% were to remove the posterior instrumentation in circumferential fusion constructs and only 2.9% involved an index LDA level.³⁵

COMPARISON TO LUMBAR FUSION

The adjacent level stress transfers from eliminating segmental motion following a lumbar fusion is of great concern, given the increased risk of symptomatic ASD necessitating additional surgery and its related morbidity and costs.³⁶ In 2021, Lau et al. published a systematic review with meta-analysis of ASD after lumbar fusion. Of 3,553 patients with a mean follow-up of 4.15 yr, the occurrence rate of ASD was 13.4% and the ASD-related surgery rate was 11.2%.³⁷ In contrast, the average occurrence rate of ASD surgery among the data presented in this article for LDA is 2.17%.

The cost of enabling technology in lumbar fusion has led to questions regarding their cost benefit ratio. In 2021, Passias et al. conducted a retrospective database study using propensity matching to compare the economic outcomes between open (n=120), minimally invasive (MIS, n=120), and robot-assisted (n=120) lumbar fusions. With a mean of 2.3 levels fused in each group, the average total cost overall was $48,019.04 ($42,538.98 open, $41,471.21 MIS, $60,047.01 robotic).³⁸ Among the data presented in this article, there is a substantial difference between costs associated with LDA compared to fusion. For fusion, the implant cost per level is over 3 times greater, the surgeon fees are over 3 times greater, and the resultant total cost is at least double compared to LDA. This does not factor in the additional costs associated with adjacent level surgery more commonly found with fusion or the cost savings associated with earlier recovery, return to work, and recreational activities following LDA.

FUTURE DIRECTIONS

As the field of motion preservation continues to advance, new artificial discs are expected to gain FDA approval along with other motion preserving implants and technologies entering the market recently and in the near future. For example, facet joint replacement devices and a single construct devices that integrate both disc and facet replacement have emerged as fusion alternatives.

For those with multilevel pathology, evidence supports hybrid surgery and 3+ levels of arthroplasty for the right patient. Many patients are denied these surgical options based on the narrow FDA guidelines. In the future, expanding the FDA guidelines would support insurance approvals and increase access to care. Updates to insurance medical policies would also increase access to care and may lead to significant cost savings. Ongoing research is needed to better define real-world applications of LDA, including when mild instability and anatomical asymmetries are present. Given the critical role of bone density in supporting artificial discs, further research should assess whether Hounsfield Unit measurements from preoperative CT scans, or other diagnostic modalities, offer a more accurate, site-specific evaluation of bone quality than DXA.

SUMMARY

There is extensive data available for both lumbar disc arthroplasty and fusion reporting short-, mid-, and long-term outcomes. In the short term, LDA and fusion have very low complications rates, but LDA has been associated with reduced blood loss, operating time, and hospital stays. Early on, the LDA group has faster improvements in patient reported outcomes, but the fusion group eventually catches up allowing both groups to experience significant improvements compared to baseline throughout the midterm. Generally, arthroplasty levels maintain segmental motion and fused levels eliminate segmental motion, which impacts global motion. Levels adjacent to fusion generally increase in motion compared to baseline, while levels adjacent to arthroplasty maintain physiological motion. In the long term, the rate of ASD is increased in the fusion group and ASD surgery occurs more frequently and at sooner intervals in the fusion group. There were no implant issues found up to 20 yr in the arthroplasty group.

CONCLUSION

In the US, artificial discs have been FDA approved for 2 decades and have over 20 yr of data to support their safety, efficacy, application in real-world settings, and favorable short- mid- and long-term outcomes. Perhaps most surprising is the lower cost associated with LDA in the short- and long-term. Despite ongoing challenges with lower reimbursement rates and insurance denials for real-world use, spine surgeons continue to champion motion preservation technologies and patients continue to be interested in preserving or restoring their motion. Although motion preservation is not an option for every patient or spinal pathology, it may one day become the gold standard treatment for refractory symptomatic DDD for the protection it provides from additional spine surgery and the undeniable health benefits of sustained or restored physical activity.