Cervical spondylotic radiculopathy (CSR) is cervical nerve root compression or irritation, causing pain, numbness, weakness, or tingling radiating from the neck to the arms and hands [1]. This condition is frequently related to cervical spine degenerative changes, including uncinate process hypertrophy, bony spur formation, facet joint degenerations, and intervertebral disc herniation with concomitant disc space narrowing, causing foraminal stenosis [2]. The primary objective in managing CSR is to alleviate nerve root compression, reduce symptoms, and restore functional capacity. Conservative treatments are frequently the initial approach, and surgical decompression by compressive structure removal is required for patients with persistent or progressive symptoms. Such interventions involve cervical disc arthroplasty (CDR), anterior cervical discectomy and fusion (ACDF), and cervical keyhole foraminotomy [3–5].

Conversely, cervical spondylotic myelopathy (CSM) is a progressive degenerative condition secondary to spinal cord compression, which significantly contributes to healthcare costs and socioeconomic burdens [6]. Typical symptoms include neck pain, heavy leg, fine hand dexterity dysfunction, upper and/or lower extremity weakness, sensory disturbances, gait instability, and falls [7]. Paralysis and bowel or bladder incontinence may be possible in more severe cases [8]. Systematic reviews have revealed that 20%–60% of patients with clinical and radiological features of CSM experience functional status deterioration in the absence of proper surgical interventions [9].

Recent developments in spinal biomechanics and advanced surgical techniques have profoundly affected the approaches in CSM management [10–12]. The surgery primarily aimed to preserve or improve neurological function, correct sagittal deformities, and ensure spinal stability. Successfully achieving these objectives is crucial for optimizing patient outcomes and ensuring favorable long-term results [10,13]. The decision-making process in selecting the appropriate surgical approach for CSM primarily depends on the patient’s clinical presentation, age, functional status, radiographic results, comorbidities, smoking history, spinal cord compression severity, and surgeon expertise in specific techniques [14–16].

A survey that included 900 case-strategy responses for assessing CSM surgical approaches revealed that 51% favored anterior surgery, 38% recommended posterior surgery, and 11% supported a combined approach. Additionally, the study emphasized the ongoing debates in cases that involve C2–C7 kyphosis with an angle of >5° or the presence of segmental kyphotic deformities [17]. Consequently, this narrative review aims to guide spine surgeons in optimizing treatment strategies for CSM and CSR by providing a stepwise, evidence-based framework for such cases.

CSR is a relatively predominant condition, with an estimated annual incidence of 85 per 100,000 individuals [1,18]. It primarily affects people aged 40–60 years, with males being more susceptible to females [1]. Risk factors for CSR include advanced age, male sex, degenerative spondylosis, trauma, heavy lifting, frequent diving, repetitive activities (e.g., prolonged overhead work), smoking, and genetic predisposition [19–21]. Conversely, Boogaart and Bartels [22] and Kokubun et al. [23] revealed that 1.6 per 100,000 and 5.7 per 100,000 individuals, respectively, received surgical treatment for symptomatic CSM [22,23]. The mean age of onset is 62.1±10.6 years, with males comprising 52.5% of cases [22]. Furthermore, 10%–15% of patients with cervical spondylosis may progress to CSM [24]. Multilevel involvement is prevalent in CSM, particularly at the C5–C6 levels caused by the high degree of mobility and susceptibility to degeneration in these segments [25,26].

CSR frequently demonstrates a positive natural course, with most patients experiencing symptom improvement without surgical intervention. Conservative management, including physical therapy, pharmacological treatment, lifestyle modifications, and interventional procedures, has achieved substantial relief in up to 90% of cases [27,28]. However, surgical decompression may be required for persistent or worsening symptoms, particularly when accompanied by neurological deficits [29].

Spinal cord compression, from static and dynamic factors, caused CSM [30]. Cervical structural abnormalities, congenital stenosis, cervical spondylosis, osteophyte formation, ossification of the posterior longitudinal ligament (OPLL), ligamentum flavum hypertrophy, kyphosis, and subluxation are static factors [31]. Dynamic factors involve spinal instability caused by compression against osteophytes or hypertrophied ligaments during flexion and extension, resulting in chronic microtrauma, ischemia, and consequent myelopathy [32].

Several studies indicated that surgery can be recommended for mild CSM cases with kyphotic curvature [15,33,34]. Patients with mild spinal cord compression where a high signal change is observed at the most affected segment on axial magnetic resonance imaging (MRI) demonstrate a significant decrease in the Japanese Orthopaedic Association (JOA) score, indicating the requirement for surgical treatment [15,35]. Patients with advanced age, longer symptom duration, higher T1 slope, and lower postoperative local kyphotic angle change may experience more gradual CSM progression and poor postoperative outcomes [36,37]. This is because chronic spinal cord compression causes neuronal loss, neuroinflammation, and vascular damage. Therefore, respective spine surgeons must educate patients with mild CSM regarding the natural disease course [38].

Electrophysiological assessments at baseline, intraoperatively, and during follow-up are crucial for detecting spinal cord dysfunction and monitoring disease progression. These are important in clinical decision-making for patients presenting with mild symptoms but demonstrate significant radiological results [39]. A study that involved the neurophysiological assessment of 70 patients with cervical spondylosis revealed that sensory evoked potentials (SEPs) were the most sensitive modality across all groups, even when compared to nerve conduction studies and electromyography [40]. Lyu et al. [41] demonstrated that among CSM cases with MRI-confirmed spinal cord signal changes, 49% demonstrated abnormalities in median nerve SEPs, whereas 47% exhibited abnormalities in tibial nerve SEPs. SEPs are particularly sensitive in determining clinically silent CSM lesions although motor evoked potentials (MEPs) are more sensitive in detecting CSM. Hence, combining SEPs with MEPs has been indicated to provide a more comprehensive approach [42].

Direct comparisons of clinical results between different surgical treatments for CSR are currently limited [28,43,44]. Luyao et al. [28] conducted a meta-analysis of six randomized controlled trials (464 patients) to identify optimal timing and treatment selection in CSR by comparing the efficacy of surgical versus conservative treatments. Their results revealed that surgical intervention caused greater reductions in neck Visual Analog Scale (VAS), arm VAS, and Neck Disability Index (NDI) scores within the first year. However, no significant differences were observed between surgical and conservative treatments in terms of range of motion (ROM) or mental health beyond 12 months. Likewise, no notable variation in clinical outcomes was found between different surgical techniques. The authors indicate that surgical intervention should be considered for patients with severe arm pain requiring symptom relief based on these results [28].

MacDowall et al. [45] conducted a propensity score-matched analysis of surgical outcomes to compare ACDF with posterior foraminotomy utilizing data from the Swedish Spine Register. Among the 4,368 patients, 647 (14.8%) underwent posterior foraminotomy, whereas 3,721 (85.2%) received ACDF. Their study demonstrated no significant differences between the two groups in terms of NDI, neck and arm pain, or adjacent segment disease incidence. The differences were not clinically significant although the posterior foraminotomy group demonstrated a higher reoperation rate [45].

Montano et al. [46] in 2019 conducted a meta-analysis of five studies to compare ACDF with laminoplasty, indicating no significant differences in terms of JOA scores, operative time, or complication rates. However, ACDF was related to reduced intraoperative blood loss and better preservation of cervical lordosis, whereas laminoplasty enabled greater ROM. The authors concluded that ACDF may be the preferred approach for managing CSM in patients with a higher compression ratio, although further randomized controlled trials with robust designs are warranted to validate these results [46].

A recent meta-analysis by Zhou et al. [47] involved 17 studies to compare the efficacy of anterior and posterior surgical approaches for multilevel CSM. Both approaches were comparable in terms of operative time, hospital stay, and JOA score. However, the anterior approach demonstrated greater effectiveness in improving NDI scores, reducing neck VAS scores, and enhancing cervical lordosis. Moreover, the anterior approach caused less intraoperative blood loss, whereas the posterior approach enabled greater ROM preservation and exhibited fewer postoperative complications [47]. Their study emphasizes distinct advantages and limitations of each approach although both approaches demonstrated favorable clinical outcomes, indicating that the choice of surgical approach should be tailored to the individual patient [47].

Luo et al. [48] conducted a systematic review and meta-analysis to evaluate clinical outcomes and complications in multilevel CSM management. No significant difference in the recovery rate was observed between the anterior and posterior surgery groups among five studies involving 420 patients. However, the postoperative complication rate was significantly higher in the anterior surgery group across nine studies involving 804 patients. Additionally, the reoperation rate was significantly higher for the anterior approach in five studies involving 294 patients. The anterior approach was related to significantly greater intraoperative blood loss and longer operative time compared to the posterior approach among four studies involving 252 patients [48]. The authors concluded no significant difference in neurological recovery rates between the two approaches, but the anterior approach may improve postoperative neurological function for multilevel CSM [48]. This indicates that surgical approach selection should be carefully considered for patients with multilevel CSM.

Yuan et al. [49] published a meta-analysis in 2019 that included 14 studies and compared the clinical safety and efficacy of laminectomy and fusion versus laminoplasty in patients with CSM. The laminectomy and fusion group demonstrated significantly higher rates for overall complications (odds ratio, 2.60; p<0.01) and root palsy (odds ratio, 3.18; p<0.01) although both groups exhibited comparable postoperative JOA scores, neck pain VAS, cervical curvature index, Nurick grade, and reoperation rates [49]. The authors concluded that laminoplasty was preferred due to lower nerve palsy rates, whereas most clinical outcomes were similar between laminectomy and fusion and laminoplasty for multilevel CSM [49].

Traynelis et al. [50] conducted a systematic review that included 10 studies and compared nontraditional surgical techniques for CSM treatment, involving CDR, oblique corpectomy, and skip laminectomy. CDR exhibited favorable neurological outcomes compared to ACDF, but functional and pain outcomes were inconsistent. Conversely, skip laminectomy achieved positive neurological outcomes in three studies, although functional and pain outcomes were inconsistently reported [50]. Currently, high-quality studies on the treatment outcomes of these techniques for symptomatic CSM are limited. Comparative prospective studies with long-term follow-up and standardized outcome measures are warranted to comprehensively assess the efficacy of nontraditional surgical approaches [50].

Liu et al. [51] involved a cohort of 180 patients with CSM to compare the outcomes of three anterior surgical techniques: ACDF, anterior corpectomy and fusion (ACCF), and hybrid decompression and fusion (HDF). Their analysis indicated no statistically significant differences between the HDF and ACDF groups in terms of clinical outcomes, cervical lordosis restoration, or postoperative complication rates. The ACCF group achieved comparable clinical outcomes to the other groups, and it was associated with greater blood loss, lower fusion rates, and higher postoperative complication incidences. Furthermore, improvements in cervical and segmental lordosis in the ACCF group were significantly less pronounced compared to the other groups, with no significant variations in adjacent-level radiographic changes among the three groups [51]. These results indicate that HDF may be a viable and safe alternative to ACDF for managing multilevel CSM, whereas ACCF presents to be a less favorable option.

Degenerative cervical myelopathy (DCM) represents a broad range of conditions characterized by spinal cord compression caused by degenerative changes in the cervical spine. These changes may include OPLL, disc degeneration, osteophyte formation, ligament hypertrophy, and spinal canal stenosis. CSM is considered the leading cause of DCM, with the latter considering other degenerative factors. In essence, CSM is a myelopathy originating from cervical spondylosis, whereas DCM represents a more comprehensive classification of myelopathy caused by any form of cervical degenerative change. The algorithm proposed by Lee et al. [52] guides determining appropriate surgical strategies in DCM.

Selecting the optimal surgical approach for CSM involves careful consideration of different factors, including cervical sagittal alignment, number of affected segments, patient condition, axial neck pain severity, and surgeon expertise (Fig. 1) [10,11]. Cervical sagittal alignment is crucial in classifying patients with cervical lordosis or kyphotic deformity using certain measurements, such as the C2–C7 sagittal vertical axis, C2–C7 lordosis, C2 and T1 slopes, and K-line [11,15,51]. The number of decompressed levels is the primary consideration for patients with maintained cervical lordosis, as alignment correction is typically unnecessary. ACDF is frequently recommended for cases involving fewer than three levels, and CDR is considered in cases with minimal disc degeneration. Laminoplasty is typically preferred for cases involving three or more levels [53]. Laminoplasty combined with foraminotomy has been a safe and effective approach in alleviating arm pain in cases of concurrent CSM and CSR, thereby minimizing risks of postoperative instability or kyphosis (Figs. 2, 3) [54].

Laminoplasty provides the advantage of preserving spinal segment mobility, maintaining posterior bony elements, and preventing postoperative kyphosis with minimal dural exposure [23,55]. In comparison, ACDF requires careful consideration because of potential implant-related complications (e.g., pseudarthrosis, graft dislodgement), reduced cervical mobility, and accelerated adjacent segment degeneration [56]. However, axial neck pain is a prevalent complication in posterior laminoplasty. Hence, laminoplasty may be less effective and should be reconsidered in cases of significant preoperative neck pain or kyphosis of >13° [57]. Meanwhile, posterior decompression and fusion are indicated when more than three levels are affected and spinal instability is present [46,48,49].

The optimal surgical approach for CSM with mild focal kyphosis remains debatable. Suda et al. [57] proposed that anterior decompression is appropriate for patients with local kyphotic angles of >13°, whereas laminoplasty is preferred for those with local kyphotic angles of <13°. The analysis by Uchida et al. [35] revealed that 61% of 43 patients with kyphotic deformities of >10° demonstrated segmental instability, and 51% had spinal canal stenosis at the kyphotic segments. Additionally, ACDF caused a significantly more increased spinal cord cross-sectional area compared to laminoplasty. Therefore, the authors emphasized the potential of correcting regional sagittal alignment in patients with kyphotic deformities of ≥10° for improved neurological recovery. Wu et al. [58] revealed that local cervical kyphosis may worsen to severe myelopathy, even in CSM without severe disc herniation. Therefore, anterior decompression with ACDF may be more favorable than laminoplasty in cases of myelopathy associated with local kyphosis (Fig. 4) [36,58].

Achieving both adequate decompression and deformity correction is crucial in CSM cases with severe kyphosis (C2–C7 kyphotic angle of >10°) [36]. Preoperative cervical traction or flexion-extension dynamic radiographs are required to assess the correctability of the deformity [59]. Posterior laminectomy and fusion are predominantly utilized for correctable severe kyphosis, whereas ACDF or vertebral body sliding osteotomy (VBSO) may be more suitable for kyphosis that involves fewer than three segments (Fig. 5) [53,60]. A combined posterior–anterior–posterior (P-A-P) approach may be necessary for severe kyphosis with significant spinal cord compression [14,61]. This includes posterior decompression and instrumentation, followed by anterior discectomy for deformity correction and final posterior fixation for cervical lordosis restoration (Figs. 6, 7).

Lee et al. [60] introduced the VBSO technique, involving intervertebral resection above and below the affected vertebral body and osteotomy medial to the uncinate process, followed by anterior advancement and fixation (Fig. 5). This technique is particularly indicated for severe kyphosis that affects fewer than three segments. A study comparing 34 VBSO cases with 31 ACCF cases indicated that VBSO caused superior C0–C2, C2–C7, and segmental lordosis correction angles compared to single-level corpectomy [60]. Thus, the selection for severe kyphosis that involves fewer than three segments should be made according to the surgeon’s preference and experience with the aforementioned techniques.

Both anterior and posterior cervical surgeries are related to distinct complications [62,63]. Posterior surgeries may cause axial pain (22.2%) and neurological deficits, such as C5 palsy (9.6%–25%), whereas anterior surgeries result in cerebrospinal fluid (CSF) leakage (2.4%–31%), dysphagia, dysarthria, and hoarseness (2.4%–16.8%) [55]. Prevalent complications regarding higher posterior cervical surgeries include postlaminectomy kyphosis, postoperative infections, iatrogenic vertebral artery injury, nerve stretching-related pain and weakness, and reduced ROM [14,64,65]. Complications for anterior cervical surgeries include vocal changes, dysphagia, CSF leakage from incidental durotomy, screw loosening, implant displacement, graft dislodgement, pseudarthrosis, pneumonia, vertebral artery injury, and neurological deficits such as quadriplegia [61,66].

Meticulous preoperative planning, including a comprehensive review of the patient’s medical history and detailed neurological assessments, is crucial to minimize these risks. Similarly, multidisciplinary discussions are essential for coordinated care. Srikandarajah et al. [67] recently introduced guidelines that outline a comprehensive care pathway for intraoperative spinal cord injury (ISCI) prevention, diagnosis, and management. The use of intraoperative neuromonitoring (IONM) is recommended in high-risk cases. The authors asserted that preoperative planning should include multidisciplinary discussions, IONM, consensus on checklists, and postoperative monitoring for patient safety. Pausing the procedure, alerting the surgical team, and ensuring all members are focused on addressing the underlying issue is imperative in the event of an ISCI. Subsequently, reversible surgical factors need to be managed by reducing traction, removing rods, confirming implant positioning, and expanding decompression areas. Assessing if factors other than surgical elements are involved, such as anesthesia and hypotension, is essential for detecting signal changes in IONM. Verification of equipment and electrode placement, retesting, anesthetic adjustment, and patient positioning reconfirmation are warranted in cases of persistent signal changes. Consideration should be given to administering steroids, performing a wake-up test, and deferring surgery with postoperative imaging if neurological deficits persist. Careful surgical and anesthetic planning, combined with rigorous intraoperative and postoperative management after the AO Spine Praxis Care Pathway, is crucial to optimize surgical outcomes and minimize complications [67].

ACDF, CDR, and posterior cervical foraminotomy are effective surgical approaches for decompressing the cervical nerve root and alleviating symptoms in patients with CSR, each providing distinct advantages and limitations. Appropriate patient selection and precise surgical execution are key to achieving optimal outcomes, considering the benign course of CSR and the comparable efficacy of these techniques.

Several factors must be considered when identifying the surgical approach for CSM, including the number of affected segments, spinal alignment, kyphosis degree, and stiffness. ACDF is viable for cases that involve fewer than three segments in patients with preserved cervical lordosis, whereas posterior laminoplasty may be suitable when more than three segments are affected. However, the surgical strategy must account for the deformity stiffness when more than three segments are affected in patients with kyphotic deformity. VBSO is an excellent technique for both deformity correction and spinal cord decompression in such cases, especially in the hands of experienced surgeons. A sequential P-A-P approach may be necessary for complex cases with severe spinal cord compression and significant deformity.

Both short-term and long-term outcomes must be considered when selecting the most appropriate surgical approach for CSR and CSM. Individualized decision-making according to patient factors and surgeon preferences is warranted, whereas an algorithmic approach can provide guidance. Thus, the algorithm proposed in this review should be regarded as a reference framework to help in achieving the best possible outcomes for patients with CSR and CSM.