Authors: Ji-Won Kwon, Kyung-Soo Suk, Seong-Hwan Moon, Si-Young Park, Namhoo Kim, Sub-Ri Park, Jae-Won Shin, Hak-Sun Kim, Byung Ho Lee
Categories: Review Article, Lumbar spinal stenosis, Biportal endoscopic spine surgery, Navigation, Robotics, Artificial intelligence
Source: Asian Spine Journal
Authors: Ji-Won Kwon, Kyung-Soo Suk, Seong-Hwan Moon, Si-Young Park, Namhoo Kim, Sub-Ri Park, Jae-Won Shin, Hak-Sun Kim, Byung Ho Lee
Lumbar spinal stenosis (LSS) is a common degenerative spinal condition where spinal canal narrowing causes symptoms such as neurogenic claudication, radiculopathy, and lower back pain. While non-operative and surgical approaches yield similar long-term outcomes, surgical intervention—particularly decompression—can provide earlier symptom relief, functional recovery, and fall prevention in selected patients with refractory symptoms. Recent advancements in surgical technologies and image guidance have brought about a paradigm shift in LSS management. Biportal endoscopic spine surgery (BESS) has gained global traction as a minimally invasive alternative to traditional decompression methods, offering superior visualization, less soft tissue damage, shorter hospital stays, and faster recovery. High-quality studies, including randomized controlled trials, have shown promising outcomes for this technique. Furthermore, the integration of navigation systems, robot-assisted instrumentation, and artificial intelligence (AI)-driven diagnostics and surgical planning tools is transforming spinal surgery by enhancing precision in preoperative evaluation and intraoperative execution. These innovations enable accurate targeting, reduce complications, and improve reproducibility across diverse surgical settings. This review provides an updated overview of LSS, covering its pathophysiology, clinical assessment, diagnosis, and treatment. Special emphasis is placed on the growing role of BESS and the transformative impact of digital technologies such as navigation, robotics, and AI in the evolving landscape of spinal stenosis care.
Lumbar spinal stenosis (LSS) is a frequently encountered degenerative spinal disorder in the elderly population. It results from progressive narrowing of the spinal canal, primarily due to intervertebral disc degeneration, facet joint hypertrophy, and buckling or thickening of the ligamentum flavum [1]. These degenerative changes lead to mechanical compression of the cauda equina and nerve roots, often manifesting clinically as neurogenic claudication, lower extremity radiculopathy, and varying degrees of axial back pain. These symptoms substantially impair patients’ quality of life and mobility [2]. For patients with mild symptoms, conservative treatment including physical therapy, analgesics, neuropathic agents, and epidural steroid injections (ESIs) may provide symptomatic relief [3]. However, for moderate-to-severe cases or those with progressive neurologic compromise, surgical decompression remains the standard of care, with numerous randomized controlled trials (RCTs) supporting its superiority over non-operative management in terms of pain relief and functional recovery [4]. Traditional open laminectomy, considered the gold standard for decompression, has significant drawbacks, including extensive paraspinal muscle dissection, increased intraoperative blood loss, postoperative pain, and prolonged recovery time, which are major concerns in the elderly population with multiple comorbidities [5]. To overcome these limitations, spinal surgery has evolved toward less invasive techniques, with biportal endoscopic spine surgery (BESS) being a promising advancement. BESS utilizes two independent working portals—one for the endoscope and one for surgical instrumentation—allowing effective neural decompression with minimal soft tissue disruption. Studies have shown that BESS achieves comparable or superior outcomes to conventional microscopic decompression, with fewer complications, less muscle injury, and shorter hospital stays [6-8]. Meanwhile, technological advancements like intraoperative navigation systems and robot-assisted pedicle screw placement are improving instrumentation accuracy, especially in complex or revision cases [9,10]. Artificial intelligence (AI) is also enhancing spinal care by enhancing imaging analysis and clinical decision-making, enabling automated detection of stenosis, risk stratification, and surgical outcome prediction, which can lead to more precise and personalized treatment [11]. Given the pace of technological innovation and the growing body of clinical evidence, this article provides an updated review of LSS, covering its pathophysiology, diagnostic evaluation, and treatment options, with a focus on the evolving role of BESS and emerging digital technologies in optimizing surgical outcomes.
The progressive narrowing of the spinal canal and neural foramina in LSS typically results from a combination of intervertebral disc degeneration, facet joint hypertrophy, and ligamentum flavum thickening or infolding [12]. These structural changes reduce the available space for neural elements, leading to mechanical compression, venous congestion, and ischemia of the cauda equina and exiting nerve roots [13]. From a biomechanical standpoint, aging triggers a degenerative cascade in the spine, involving annular fissuring, disc space collapse, and vertebral instability. This leads to compensatory hypertrophy of facet joints and the ligamentum flavum. Loss of disc height reduces lumbar lordosis (LL) and contributes to relative kyphosis, increasing stress on the posterior spinal elements and further exacerbating central canal narrowing [14]. The ligamentum flavum, normally tense in extension, becomes redundant and folds inward with disc height loss, significantly contributing to central stenosis [15]. These pathoanatomic changes often culminate in the hallmark symptom of neurogenic claudication— exertion-induced lower extremity pain relieved by rest or lumbar flexion [16]. LSS can be anatomically classified based on the site of neural compression into central canal stenosis, lateral recess stenosis, foraminal stenosis, and extraforaminal (far-out zone) stenosis [17]. Pathologically, stenosis can result from congenital abnormalities (e.g., short pedicles or achondroplasia), degenerative spondylosis, inflammatory arthropathies, or metabolic disorders like Paget’s disease or fluorosis [18]. The anatomical and pathological classification of LSS is summarized in Table 1. Importantly, the severity of radiographic stenosis does not always correlate with symptom intensity, a phenomenon attributed to dynamic components such as posture-induced mechanical loading, transient vascular compromise, and altered pain processing involving central sensitization [19]. Contemporary understanding of LSS recognizes its dynamic nature, with symptom exacerbation often occurring during lumbar extension (which narrows the canal) and improvement during flexion (which enlarges it) [20] (Table 1). This understanding highlights the importance of dynamic imaging, such as upright or axial-loaded magnetic resonance imaging (MRI) and supports minimally invasive decompression strategies that prioritize neural element preservation while minimizing iatrogenic destabilization.
LSS prevalence increases with age, and it is a leading cause of pain and disability in adults aged over 60 [21]. While radiological evidence of spinal stenosis is common in those over 70 (up to 78% [22]), estimates of symptomatic prevalence range between 11% and 38%, depending on the diagnostic criteria and modalities, with a higher incidence in women than men [23,24]. Population-based studies using administrative databases (e.g., the 10th revision of the International Classification of Diseases codes), imaging, and clinical assessments suggest that LSS is significantly underdiagnosed in its early stages, partly because initial symptoms, such as intermittent leg pain or fatigue during walking, may be attributed to other musculoskeletal or vascular conditions like peripheral arterial disease or hip osteoarthritis [25]. Moreover, MRI frequently reveals moderate-to-severe stenosis in asymptomatic individuals, particularly in the elderly, highlighting the imperfect correlation between imaging severity and clinical presentation [26,27]. The natural history of LSS is highly variable. Some patients experience a gradual worsening of symptoms, while others remain stable or show spontaneous improvement, particularly with non-surgical management. Longitudinal cohort studies show that approximately 50% of conservatively-treated patients remain stable or improve over 4–10 years [28]. However, 30%–40% of these patients eventually require surgery due to worsening neurogenic claudication, radiculopathy, or functional decline [29]. Surgical intervention, particularly decompression with or without fusion, has shown superior outcomes in terms of pain relief, walking capacity, and quality of life compared to continued non-operative care. In the landmark Spine Patient Outcomes Research Trial, surgically treated patients showed significantly greater improvement in leg pain, Oswestry Disability Index (ODI), 36-Item Short Form Survey physical function, and satisfaction scores at 2- and 4-year follow-ups [4]. The burden of LSS is expected to rise due to the progressive population aging and increasing demand for mobility-preserving interventions. In this context, understanding the epidemiologic trends and variable natural course of LSS is crucial for guiding individualized clinical decision-making and informing public health strategies and resource allocation in spine care systems worldwide.
LSS typically presents with a characteristic constellation of symptoms, including neurogenic claudication, lower extremity radiculopathy, and chronic lower back pain [2]. Neurogenic claudication—often described by patients as bilateral or unilateral leg pain, numbness, tingling, or heaviness worsened by standing or walking and relieved by lumbar flexion or sitting—is the hallmark symptom of LSS [30]. Many patients report decreased walking distance and adopt a forward-flexed posture while ambulating, sometimes using support like shopping carts or railings, a phenomenon referred to as the “shopping cart sign” [31]. Pain distribution varies with stenosis central canal stenosis typically causes bilateral, often symmetric leg symptoms associated with claudication, while foraminal or lateral recess stenosis leads to radicular pain following a dermatomal distribution [17]. In more severe cases, symptoms may progress to gait disturbance, motor weakness, or rare but concerning symptoms like bladder and bowel dysfunction, suggesting cauda equina syndrome [32]. Clinical evaluation involves a comprehensive history and physical examination, including neurologic testing of strength, deep tendon reflexes, and dermatomal sensory function. Provocative maneuvers like the walking test, two-stage treadmill test, and extension-based exercises (e.g., standing back extension) can help reproduce symptoms and differentiate neurogenic claudication from vascular etiologies [33]. To rule out peripheral arterial disease, palpation of distal pulses and measurement of ankle-brachial index (ABI) are recommended. An ABI <0.9 indicates arterial insufficiency, warranting vascular evaluation [34]. Standardized clinical criteria have been proposed to improve diagnostic accuracy. The N-CLASS (Neurological Claudication caused by LSS) criteria emphasize the presence (1) leg or buttock symptoms during walking or standing, (2) relief with lumbar flexion, (3) motor or sensory changes in the lower extremities, and (4) intact dorsalis pedis pulses to rule out vascular claudication [24]. Additionally, physical performance measures such as grip strength, Timed Up and Go test, and Functional Mobility Scale are objective correlates of symptom severity and risk of falls in LSS populations, particularly in elderly individuals [35,36]. Differential diagnosis is crucial in elderly patients, as they often present with coexisting musculoskeletal and neurologic disorders like diabetic peripheral neuropathy, hip or knee osteoarthritis, sacroiliac joint dysfunction, and vascular claudication. A comprehensive approach, integrating patient history, focused examination, and functional assessments, is necessary for accurate diagnosis. Because patient-reported symptoms and imaging findings do not always correlate perfectly, clinicians must interpret symptoms in the context of neurologic deficits, gait impairment, and functional limitation to guide management strategies [26].
Evaluating LSS requires combining clinical judgment with multimodal imaging to determine the anatomical level, severity, and surgical indications. Recent technological advancements, including navigation systems, robotic platforms, and AI, have expanded the diagnostic and therapeutic landscape.
Standard lumbar radiographs are a fundamental firstline imaging modality for LSS. Lateral views assess segmental instability, disc space narrowing, and anterolisthesis or retrolisthesis, while anteroposterior views reveal facet joint hypertrophy, scoliotic curvature, and osteophyte formation [37]. Dynamic flexion-extension radiographs evaluate segmental instability, with translational movement >4 mm or angular motion >10°–15° indicating dynamic instability requiring fusion [38]. Plain radiographs also help evaluate global spinopelvic alignment parameters, including pelvic incidence, pelvic tilt, LL, and sagittal vertical axis. These parameters are important for predicting functional outcomes, postoperative satisfaction, and fall risk, particularly in elderly LSS patients [39,40].
MRI is the gold standard for evaluating spinal stenosis due to its excellent soft tissue contrast and ability to assess neural structures [41]. T2-weighted sagittal and axial images provide excellent visualization of thecal sac compression, nerve root impingement, ligamentum flavum hypertrophy, facet joint osteoarthritis, and disc herniation or bulging. T1-weighted images complement this by delineating fat content and help detect foraminal stenosis, fatty infiltration of musculature, and vertebral body changes. Recent advancements, including axial-loaded MRI and upright (weight-bearing) MRI, can better correlate with symptom severity by simulating physiologic loading, improving diagnostic sensitivity for dynamic stenosis, but are currently limited by availability and cost [20]. Notably, the radiological severity of stenosis on MRI does not always correlate with symptom burden. Several studies have documented moderate-to-severe stenosis in asymptomatic individuals, especially the elderly. Therefore, integrating MRI findings with patient history and physical examination is essential for accurate diagnosis [27].
Computed tomography (CT) scanning provides superior resolution of osseous structures, particularly valuable in preoperative planning when bony abnormalities dominate the pathology. CT is particularly effective in identifying osteophytes, facet joint overgrowth, calcified disc herniation, and fusion mass status—areas where MRI may be less sensitive [42]. CT serves as an alternative for patients who cannot undergo MRI due to implanted devices, metallic implants, or claustrophobia. Multiplanar reconstructions in sagittal and axial views further enhance the assessment of canal and foraminal dimensions [43].
CT myelography, though less commonly used, is useful in specific clinical contexts such as postoperative patients with persistent symptoms, multilevel stenosis with equivocal MRI findings, or cases with extensive hardware-induced artifacts [44]. The technique involves the injection of intrathecal contrast, followed by CT acquisition, offering real-time visualization of the subarachnoid space and nerve root sleeves. It remains the most sensitive modality for detecting compressive pathology when MRI is nondiagnostic or contraindicated. However, it is an invasive modality associated with the risks of post-dural puncture headache, infection, or contrast reactions, requiring careful patient selection.
Electrodiagnostic evaluations like nerve conduction studies and needle electromyography (EMG) are useful adjuncts when imaging results are inconclusive or when differential diagnosis includes other neurologic conditions such as diabetic polyneuropathy, motor neuron disease, or peripheral nerve entrapments [45]. EMG can help confirm chronic radiculopathy, detect denervation potentials, and assess reinnervation, providing prognostic insights. Although not routinely employed as a first-line test, EMG may be helpful in multilevel or ambiguous cases, supporting diagnosis and guiding surgical planning.
Image-guided navigation systems, often integrated with intraoperative three-dimensional imaging such as the O-arm or C-arm, have become widely adopted in complex spinal procedures, including those for LSS. These technologies provide real-time spatial localization, enabling precise identification of surgical levels and facilitating accurate pedicle screw placement, especially in minimally invasive settings where anatomic landmarks are obscured [10]. Studies have shown that navigation-assisted techniques lead to higher instrumentation accuracy, lower screw misplacement rates, and reduced intraoperative radiation exposure for both the surgical team and patient [46,47]. Intraoperative navigation also enhances safety in obese patients or those with anatomical distortion from previous surgeries. It supports muscle-sparing approaches such as minimally invasive transforaminal lumbar interbody fusion (MI-TLIF) and biportal endoscopic-assisted fusion, where limited visualization demands high surgical accuracy [48] (Fig. 1).
Robotic-assisted platforms—including Mazor X (Medtronic, Minneapolis, MN, USA), ExcelsiusGPS (Globus Medical, Audubon, PA, USA), and ROSA Spine (Zimmer Biomet, Warsaw, IN, USA)—have further enhanced intraoperative precision by integrating preoperative imaging with robotic arm guidance, facilitating automated trajectory planning and intraoperative execution [9]. In the context of LSS with segmental instability, deformity, or multilevel disease, robotics help ensure consistent screw placement and alignment correction, while minimizing human error. Robot-assisted fusion surgeries have been shown to improve screw placement accuracy, reduce revision rates due to hardware failure, and enhance safety, particularly for less experienced surgeons or in anatomically challenging cases [49,50]. Robotics also reduces operative variability and improves efficiency in long-segment fusions, such as those for degenerative scoliosis or spondylolisthesis-associated stenosis.
AI-driven technologies are rapidly emerging in spinal disorder diagnostic imaging and surgical decision-making. Deep learning-based algorithms can automate vertebral structure segmentation on CT and MRI, accurately grading central canal narrowing, lateral recess compromise, and foraminal stenosis, with diagnostic accuracy comparable to expert radiologists [11]. Convolutional neural networks (CNNs) have shown promise in classifying stenosis severity with high accuracy, achieving area under the curve values exceeding 0.90 [51]. AI models are also being used to predict surgical outcomes, such as fusion success, risk of reoperation, and postoperative functional recovery, by integrating clinical, radiological, and biomechanical data. Additionally, AI-integrated robotic platforms can suggest optimal screw trajectories, cage size, and angulation, enhancing preoperative planning efficiency and personalization. While these tools are still evolving, their ability to reduce interobserver variability, enhance standardization, and offer predictive analytics represents a significant step toward precision spine surgery.
LSS treatment involves a stepwise, patient-centered approach, starting with conservative strategies and progressing to surgical intervention when necessary. Clinical decisions are based on symptom severity, duration, functional impairment, radiographic findings, and comorbidities. Despite advancements in surgical techniques, a thorough understanding of pharmacological, interventional, and operative modalities remains crucial for effective treatment.
Pharmacological treatment is the first-line therapy for LSS, particularly in patients with mild-to-moderate symptoms without neurological deficits [1]. Nonsteroidal anti-inflammatory drugs are widely used to reduce inflammation and nociceptive pain arising from facet arthropathy and soft tissue strain. Selective COX-2 inhibitors are often preferred in elderly patients due to fewer gastrointestinal side effects. For radicular pain or neurogenic claudication, neuropathic pain modulators such as gabapentin and pregabalin are often prescribed [52]. In a randomized controlled trial by Kim et al. [53], a combination of limaprost and pregabalin was found to significantly improve walking distance and leg pain compared to monotherapy, suggesting synergistic benefits of vasodilation and neuropathic pain control. Antidepressants, particularly serotonin-norepinephrine reuptake inhibitors (e.g., duloxetine), may benefit patients with comorbid depression or centralized pain syndromes, though direct evidence is limited. Muscle relaxants and oral corticosteroids can provide shortterm relief in acute exacerbations, but long-term use is discouraged due to risks of sedation, dependence, and systemic side effects. In elderly patients, careful monitoring is necessary due to the risk of polypharmacy, altered drug metabolism, and increased sensitivity to central nervous system effects. Non-pharmacologic modalities, such as physical therapy, are essential adjuncts in this population [28].
ESIs—including interlaminar, transforaminal, and caudal approaches—are commonly used as intermediate treatments for patients who are unresponsive to medications but are not yet surgical candidates [16,31]. The primary mechanism involves local corticosteroid-mediated suppression of inflammation and edema around impinged nerve roots. Interlaminar ESIs target the posterior epidural space and are particularly useful in central canal stenosis. Transforaminal injections allow precise targeting of nerve roots in lateral recess or foraminal stenosis. Caudal blocks are preferred in challenging cases, such as prior surgery or obesity, allowing medication to diffuse to the lumbosacral space. Several RCTs and meta-analyses have confirmed the short-term efficacy of ESIs in reducing pain and improving walking capacity. A multicenter RCT by Friedly et al. [54] showed benefits up to 3 months with steroids or local anesthetics, but effects were not sustained for 6 months. Caudal epidural blocks have also been validated in RCTs. Emerging interventional modalities such as epidural neuroplasty (adhesiolysis) and balloon decompression show promise for refractory cases, but more research is needed to confirm long-term efficacy and safety. Despite their utility, ESIs do not modify the natural history of LSS and should be considered adjunctive or bridging therapies. Repeated injections beyond 3 times per year are generally discouraged due to diminishing benefits and increased risk of systemic adverse effects [55]. However, ESIs performed >2 weeks prior to spinal surgery do not appear to increase the risk of postoperative infection [56].
Surgical treatment is indicated in patients with persistent symptoms despite 3–6 months of conservative care or in those presenting with progressive neurological deficits, cauda equina syndrome, or intolerable neurogenic claudication impairing quality of life. The primary objective is neural decompression, with spinal stabilization considered if instability is present. Surgical decompression can also reduce the risk of falling in LSS patients compared to non-surgical treatment [57].
Decompression without fusion is suitable for patients with isolated central or lateral recess stenosis, preserved segmental stability, and no significant spinal deformity [2]. Techniques include open laminectomy, microscopic laminotomy, tubular minimally invasive decompression, and endoscopic decompression, with the choice depending on the surgeon’s experience and disease extent. The optimal extent of decompression in LSS surgery is debated. Overly aggressive facet joint resection (>50%) may cause iatrogenic instability and increase the risk of postoperative spondylolisthesis or need for fusion [58]. Conversely, inadequate decompression can lead to persistent symptoms or early revision surgery, making precision in bone and soft tissue removal essential. For this reason, a facet-preserving strategy, especially when utilizing microscopic or endoscopic techniques, is now favored, particularly for single-level disease [59]. In patients with multilevel stenosis and preserved sagittal alignment, decompression alone may yield favorable outcomes [60]. However, caution is warranted in elderly patients, especially those with subtle or occult dynamic instability, as traditional flexion-extension radiographs may not detect segmental motion that could compromise surgical durability.
Each of the fusion techniques for surgical management of LSS has evolved from distinct historical and biomechanical frameworks, with their advantages and limitations. Posterolateral fusion (PLF) is a well-established lumbar fusion technique that has been commonly used since the mid-20th century [61]. It involves decortication of the transverse processes and placement of autograft or allograft material in the posterolateral gutters. PLF is a relatively less invasive option that avoids anterior or central spinal column dissection, often used in conjunction with decompressive laminectomy in patients with bilateral symptoms or multilevel disease. However, PLF does not directly restore disc height or foraminal volume, and fusion rates, especially without instrumentation, are typically lower compared to interbody fusion methods. Additionally, lack of anterior column support may increase the risk of pseudoarthrosis in osteoporotic or high-stress segments. Transforaminal lumbar interbody fusion (TLIF) was first introduced by Drs. Harms and Rolinger in 1982 as a modification of posterior lumbar interbody fusion (PLIF), with the intent to reduce neural manipulation and dural traction [62]. TLIF reduces neural manipulation and dural traction by accessing the disc space through a unilateral facetectomy, allowing for interbody cage placement with autograft or bone substitute. This technique offers several advantages, including restoration of disc height and foraminal space, provision of anterior column support through the interbody cage, and high fusion rates, especially with posterior instrumentation. It can be performed via open or minimally invasive approaches, with the latter reducing blood loss, muscle disruption, and hospital stay. However, limitations include the learning curve for minimally invasive TLIF, potential cage migration, and in some cases, incomplete decompression of the contralateral side unless bilaterally addressed.
Oblique lumbar interbody fusion (OLIF) is a more recent innovation, first described by Mayer [63] in 1997 and later refined in subsequent clinical series [64]. OLIF accesses the disc space via an anterior-to-psoas retroperitoneal approach, preserving posterior elements and reducing the risk of psoas muscle or lumbar plexus injury, often associated with direct lateral interbody fusion (DLIF)/extreme lateral interbody fusion (XLIF). OLIF enables the placement of large, lordotic cages, providing strong anterior column support and indirect decompression. It is particularly useful in multilevel or revision surgeries, offering favorable sagittal alignment correction and reducing postoperative fall risk [65]. However, it has a steep learning curve and carries risks of vascular injury and ureteral damage, especially in patients with variant anatomy. Anterior lumbar interbody fusion (ALIF) is among the oldest anterior approaches, with roots in general surgical access to the retroperitoneum dating back to the 1930s [66]. Its application to spine surgery significantly advanced in the 1980s and 1990s with the introduction of interbody cages and anterior plating systems. ALIF provides direct access to the disc space without disrupting posterior paraspinal musculature, allowing thorough discectomy, endplate preparation, and insertion of large, lordotic cages. It is most commonly used at the L5–S1 level due to the relatively favorable vascular anatomy. ALIF is highly effective in restoring LL and is often used as part of global sagittal balance correction strategies. However, it requires access surgeons and carries risks of iliac vessel injury, retrograde ejaculation (due to superior hypogastric plexus injury), and visceral or ureteral damage [67]. Moreover, ALIF is typically not feasible above L4–L5 due to the bifurcation of the great vessels.
The choice of fusion technique depends on patient anatomy, pathology, and surgical objectives. TLIF and PLF are suitable for posterior-based decompression and fusion, while OLIF and ALIF offer robust anterior column correction with minimal posterior disruption. Ultimately, the choice of technique depends on the level involved, clinical goals, the surgeon’s experience, and available institutional resources.
Minimally invasive spine surgery (MISS) has gained popularity for managing lumbar degenerative diseases, including LSS, particularly in patients requiring decompression with or without interbody fusion. Techniques such as MIS-TLIF, XLIF/DLIF, and biportal endoscopic fusion offer benefits such as reduced soft tissue trauma, less blood loss, lower infection rates, and shorter hospital stays [36,63,66]. RCTs and meta-analyses have shown MISS to be non-inferior to open surgery in terms of fusion rates, functional outcomes, and pain control. In a prospective randomized trial, Miller et al. [68], patients undergoing minimally invasive TLIF for low-grade spondylolisthesis demonstrated similar improvements in ODI scores and fusion rates at 1 year compared to those undergoing standard TLIF [68]. However, the minimally invasive group experienced significantly shorter postoperative hospital stays, indicating reduced perioperative morbidity without compromising clinical or radiological outcomes. Furthermore, there were no significant differences in radiological fusion rate at 12 months between the two groups. Similarly, a meta-analysis (over 1,100 patients) found MI-TLIF achieved comparable fusion rates to open TLIF with significantly less blood loss and lower postoperative infection risk [69]. However, MISS has limitations, including a steep learning curve, limited visualization, and potential for incomplete contralateral decompression unless bilateral approaches are used. In deformity cases or multilevel disease, MISS may be insufficient for correcting global alignment without navigation support or advanced planning technologies. However, advances in navigation, robotics, and endoscopic optics are addressing these challenges, solidifying MISS as a core component of modern spine surgery.
The shift toward minimally invasive surgical strategies has revolutionized the treatment paradigm for LSS, driven by the need to reduce tissue damage, enhance postoperative recovery, and improve outcomes in elderly patients. Among various minimally invasive techniques, BESS has gained substantial international traction and is now one of the most rapidly evolving spinal decompression modalities [6]. BESS employs two independent working portals for the endoscope and instrumentation, enabling triangulated manipulation and wide-angled visualization [8]. In contrast to conventional microscopic decompression, which offers a linear view, BESS provides a panoramic endoscopic view under continuous saline irrigation, facilitating precise dissection and safe removal of hypertrophied ligamentum flavum, facet overgrowth, and laminar spurs [70]. Importantly, the approach also enables preservation of posterior elements—such as the paraspinal musculature, interspinous ligaments, and facet joints—which are often compromised in open laminectomy or even microscopic laminotomy. Several studies have demonstrated the effectiveness of BESS in treating LSS. In an RCT, Park et al. [71] compared BESS with microscopic decompression for patients with central and lateral recess stenosis. The BESS group had significantly less early postoperative pain, faster ambulation, and shorter hospital stays, with comparable long-term functional outcomes (ODI and Visual Analog Scale [VAS] scores) [71]. Beyond decompression, BESS is increasingly being utilized in fusion procedures, particularly BESS-TLIF. Early reports indicate that BESS-TLIF may cause less paraspinal muscle atrophy and lower serum creatine kinase levels compared to MI-TLIF, suggesting reduced muscular trauma. In a prospective cohort study by Park et al. [7], patients undergoing BESSTLIF exhibited significant postoperative improvements in ODI and VAS scores. The reported fusion rate at 12 months was comparable to that of traditional MI-TLIF, further underlining its effectiveness [7]. Despite its rising popularity, BESS is operator-dependent with a steep learning curve, especially in complex cases like high-grade foraminal stenosis or severe epidural scarring. However, dedicated training programs, cadaver-based simulation labs, and digital navigation platforms are helping mitigate these challenges, driving broader adoption in Asia, Europe, and North America [72]. Notably, BESS is highly compatible with emerging technologies such as intraoperative navigation and robotic systems. Navigation-integrated BESS platforms are being developed to provide real-time anatomical guidance during decompression and increase safety in complex or revision surgeries [73]. Additionally, AI-enhanced preoperative planning systems are being explored in conjunction with BESS, allowing for personalized targeting, risk prediction, and automated segmentation, reinforcing its role at the intersection between MISS and digital medicine [73,74]. In summary, BESS has progressed from an experimental endoscopic technique to a validated surgical option for decompression and fusion in LSS, supported by a growing body of high-quality clinical evidence. Its adaptability and synergy with other cutting-edge technologies position BESS at the forefront of contemporary MISS (Figs. 2, 3).
Integration of digital technologies like navigation systems, robotics, and AI has transformed surgical practice, including spine surgery, over the past decade. These tools are being increasingly used in treating LSS, where precise decompression and accurate instrumentation are vital for maximizing patient outcomes while minimizing surgical complications [75]. Navigation-assisted spine surgery has gained widespread acceptance, particularly for pedicle screw placement. Intraoperative systems—whether fluoroscopy-based or CT-based—provide real-time guidance, enhancing accuracy of screw trajectory, implant positioning, and safe decompression margins [47]. Studies have consistently shown that navigated techniques outperform traditional freehand approaches. A multicenter prospective cohort study found O-arm–based navigation reduced pedicle breach rate from 10.6% to 2.3% and lowered radiation exposure for the surgical team [46]. Robotic platforms like Mazor X (Medtronic), ExcelsiusGPS (Globus Medical), and ROSA Spine (Zimmer Biomet) further enhance precision by integrating preoperative CT with intraoperative fluoroscopy for patient-specific screw trajectories [49]. Robotic-guided procedures for LSS decompression and fusion have demonstrated high screw accuracy, reduced surgeon variability, and lower reoperation rates. A meta-analysis of over 2,000 patients found robotic assistance significantly improved screw placement accuracy and reduced complications such as dural tears and neurological deficits [47]. Importantly, these technologies are not limited to open or MI-TLIF but are now being explored in endoscopy-assisted procedures. Early pilot studies indicate the feasibility and safety of robot- and navigation-assisted biportal endoscopic TLIF, demonstrating precise cage placement and effective decompression under direct endoscopic visualization [76]. Although AI is still in early clinical adoption, it shows promise in diagnostics, risk stratification, and surgical planning. For example, deep learning algorithms can automatically detect and quantify LSS on MRI with accuracy comparable to expert radiologists. A CNN model developed by Jamaludin et al. [11] achieved over 90% sensitivity in detecting central canal and foraminal stenosis on T2-weighted axial MRI. AI-based preoperative prediction models are also emerging to forecast fusion success, postoperative complications, and functional outcomes based on clinical, radiographic, and intraoperative data. Integrating navigation, robotics, and AI within a single digital surgical ecosystem marks a shift toward data-driven, patient-specific spine care. Particularly in LSS, where anatomical variability, comorbidity burden, and surgical precision intersect, these tools can improve preoperative planning, intraoperative safety, and postoperative care. Despite barriers like cost, accessibility, and training, the growing evidence supports their potential integration into routine spinal surgical practice [48,51].
As spinal surgery advances with technology, LSS management is poised for further evolution. Current evidence supports the efficacy and safety of BESS and other minimally invasive techniques. Future research will likely refine indications, enhance patient selection, and seamlessly integrate digital platforms [74]. A promising area is combining endoscopic techniques with real-time navigation and robotic systems. Preliminary pilot studies suggest that integrating robotic trajectory planning with endoscopic decompression or interbody fusion could improve implant accuracy while maintaining endoscopy’s tissue-sparing benefits [36,74]. This hybrid approach may be particularly beneficial in complex anatomies or revision cases where normal landmarks are disrupted. AI is also expected to play a major role in standardizing diagnostics and surgical planning. While existing AI models focus on automated MRI grading, future AI models may integrate multimodal data, such as gait analysis, EMG, patient-reported outcomes, and genomic profiles, to create personalized treatment algorithms [11]. These tools can help surgeons choose the optimal surgical approach, estimate complication risks, and predict functional recovery trajectories. Customizing care for elderly and frail patients with LSS is crucial, as they often have sarcopenia, frailty, and comorbidities, requiring personalized operative planning. Minimally invasive procedures like BESS—especially when combined with intraoperative neuromonitoring and image-guided navigation—may help minimize morbidity and accelerate recovery in these high-risk patients [77]. However, there is a need for more robust prospective data on outcomes in octogenarians or medically complex cohorts. As endoscopic surgery expands globally, education and standardization efforts must keep pace. This includes developing validated training curricula, cadaver-based simulations, and international consensus guidelines on indications, contraindications, and outcomes reporting [72]. Setting proficiency benchmarks and credentialing standards will help maintain uniform standards of care. Lastly, large-scale, multicenter RCTs directly comparing BESS, MI-TLIF, navigation-augmented fusion, and traditional open surgery would also advance the field. While current evidence supports the clinical utility of BESS, variations in technique, instrumentation, and surgeon experience limit its generalizability. Future studies should examine cost-effectiveness, return-to-work rates, patient-reported quality of life, and long-term reoperation risks, which are crucial in value-based care models.
LSS is a complex degenerative disorder characterized by progressive neural compression and functional decline, particularly in the elderly. Its clinical presentation varies and does not always correlate with imaging severity, requiring individualized diagnosis and treatment. Conservative treatment, including pharmacological therapy, structured rehabilitation, and ESIs, is often the first-line approach. For those who do not respond or have progressive neurologic compromise, surgical decompression with or without fusion has consistently shown superior outcomes in terms of pain relief, walking capacity, and quality of life. The choice between decompression alone and decompression with fusion must be tailored based on the presence of segmental instability, sagittal imbalance, and patient-specific factors such as age, frailty, and functional demands. Recent advancements in MISS, such as BESS, have altered the surgical landscape. BESS offers a less invasive yet effective alternative to conventional open or microscopic surgery, supported by growing high-level evidence across diverse clinical scenarios, including decompression and fusion. In parallel, the integration of navigation systems, robotic assistance, and AI is enhancing spinal care by enabling more precise, reproducible, and personalized surgical planning and execution.