Between January 2013 and December 2022, a total of 753 cases of CSM patients underwent single-segment ACDF surgery at the Spine Surgery Department of the Affiliated Hospital of Qingdao University. After applying the specified inclusion and exclusion criteria, our study focused on 65 patients (38 males, 27 females). The average age of these patients was 52.23 ± 10.8 years, ranging from 27 to 73 years. The surgical segments and corresponding case distribution were as follows: C2–3 segment (10 cases), C3–4 segment (12 cases), C4–5 segment (16 cases), C5–6 segment (14 cases), and C6–7 segment (13 cases). All patients presented symptoms of cervical spinal cord compression, including neck and shoulder discomfort, as well as numbness in one or both upper limbs, exacerbated by activity and relieved by rest. Additionally, 26 cases (40%) exhibited walking instability characterized by unsteady gait and a sensation of walking on cotton. Anteroposterior and lateral radiographs of the cervical spine revealed a reduction in the height of the intervertebral space at the affected segment. The CT and MRI scans indicated cervical disk herniation, thickening of the posterior longitudinal ligament, and ossification of the posterior longitudinal ligament. Notably, 20 patients (30.77%) exhibited signs of spinal cord degeneration.

Clinical efficacy

In the final follow-up assessment of the 65 patients, 40 cases (61.54%) were categorized as excellent, 23 cases (35.38%) as good, and 2 cases (3.08%) as average, based on the Odom criteria.

Consistency test results

Due to the extensive volume of measurement data, illustrative examples were extracted from the preoperative measurements of AxCSA, RLS, and CESA at the C2-3 level. The ICC values for AxCSA, RLS, and CESA, between the two measurers, were 0.976 (95% CI 0.962, 0.986), 0.993 (95% CI 0.989, 0.996), and 0.975 (95% CI 0.960, 0.985), respectively. The outcomes of the consistency test demonstrated that all ICC values exceeded 0.75, signifying a high level of agreement between the measurements conducted by the two observers.

Changes of AxCSA and CESA after surgery

There are significant statistical differences in the AxCSA of the operative segment at various preoperative and postoperative follow-up time points (F = 832.2, P < 0.0001). The AxCSA of the operative segment exhibited a consistent reduction in postoperative measurements at the 3rd month, 12th month, and the last follow-up compared to preoperative values (141.62 ± 19.78 mm2). The postoperative AxCSA data at these respective time points were (119.42 ± 20.08) mm2 (P < 0.001), (117.59 ± 19.69) mm2 (P < 0.001), and (117.41 ± 19.19) mm2 (P < 0.001), indicating statistically significant differences. The postoperative AxCSA measurements at these intervals reflected a decrease of 15.68%, 16.97%, and 17.10% in comparison with preoperative data. Conversely, for the AxCSA of the non-operative segment, no significant differences were observed between the preoperative data and the postoperative data obtained at each follow-up time point (P > 0.05). Additionally, there were no significant differences in the AxCSA for the non-operative segment between preoperative and postoperative follow-up time points (F = 0.8159, P = 0.4641).

In the context of the AxCSA analysis pertaining to the C2-3 vertebral segment, the measurement results at various preoperative and postoperative follow-up time points showed significant statistical differences (F = 264.3, P < 0.0001). The preoperative data exhibited a mean value of (131.81 ± 22.88) mm2, subsequent to the surgical intervention, a discernible reduction was observed in the postoperative measurements, with values recorded as (104.88 ± 19.57) mm2 (P < 0.001), (105.86 ± 21.27) mm2 (P < 0.001), and (107.01 ± 20.45) mm2 (P < 0.001) for the respective analyses, as illustrated in Fig. 2. The observed disparities between preoperative and postoperative AxCSA measurements were determined to be statistically significant (Fig. 2).

Fig. 2

Schematic diagram of the changes in the cross-sectional area of the cervical longus (AxCSA) at the surgical segment before and after surgery. The data show a significant decrease in AxCSA at each surgical segment after surgery, which indicates that the cervical longus underwent a certain extent of atrophy after ACDF surgery

In the context of the AxCSA analysis related to the C3–4 vertebral segment, the measurement results at various preoperative and postoperative follow-up time points showed significant statistical differences (F = 186.4, P < 0.0001). The preoperative data demonstrated a mean value of (137.86 ± 18.21) mm2, following the surgical intervention, a noteworthy reduction in postoperative measurements was observed, registering as (118.39 ± 17.64) mm2 (P < 0.001), (115.80 ± 17.71) mm2 (P < 0.001), and (113.13 ± 17.58) mm2 (P < 0.001) for the respective analyses, as delineated in Fig. 2. These observed discrepancies between preoperative and postoperative AxCSA measurements were found to be statistically significant (Fig. 2).

In the examination of AxCSA within the C4–5 vertebral segment, the measurement results at various preoperative and postoperative follow-up time points showed significant statistical differences (F = 482.3, P < 0.0001). The preoperative data revealed a mean value of (140.41 ± 19.24) mm2, following the surgical intervention, a discernible reduction in postoperative measurements was evident, registering at (117.40 ± 17.32) mm2 (P < 0.001), (113.42 ± 16.23) mm2 (P < 0.001), and (115.31 ± 17.16) mm2 (P < 0.001) for the respective assessments, as depicted in Fig. 2. These variations between preoperative and postoperative AxCSA values were determined to be statistically significant, as illustrated in Fig. 2.

In the context of the AxCSA analysis pertaining to the C5–6 spinal segment, the measurement results at various preoperative and postoperative follow-up time points showed significant statistical differences (F = 194.7, P < 0.0001). The preoperative measurements revealed a mean value of (144.50 ± 17.46) mm2, following the surgical intervention, the corresponding postoperative measurements exhibited a discernible reduction to (122.62 ± 20.56) mm2 (P < 0.001), (123.66 ± 20.46) mm2 (P < 0.001), and (125.81 ± 20.51) mm2 (P < 0.001), as illustrated in Fig. 2. These alterations in postoperative measurements achieved statistical significance, as denoted in Fig. 2.

In the analysis of the AxCSA pertaining to the C6–7 segment, the measurement results at various preoperative and postoperative follow-up time points showed significant statistical differences (F = 244.0, P < 0.0001). The preoperative measurement yielded (151.00 ± 19.88) mm2, subsequent to the surgical intervention, the postoperative assessments revealed a notable reduction to (130.56 ± 20.42) mm2 (P < 0.001), (126.84 ± 19.52) mm2 (P < 0.001), and (122.91 ± 17.70) mm2 (P < 0.001), as depicted in Fig. 2. Importantly, these postoperative alterations in AxCSA values were found to be statistically significant, as underscored in Fig. 2.

In the analysis of CESA within both operative and non-operative segments, no statistically significant distinctions were observed when comparing preoperative data with postoperative data at the 3rd month, 12th month, and the final follow-up (P > 0.05). Meanwhile, there were no significant differences in the CESA between preoperative and postoperative follow-up time points for the operative segment (F = 2.506, P = 0.1157), and no significant differences were observed in CESA for the non-operative segment between preoperative and postoperative follow-up time points (F = 1.003, P = 0.3567).

Postoperative changes of RLS

There are significant statistical differences in the RLS of the operative segment at various preoperative and postoperative follow-up time points (F = 513.4, P < 0.0001). In the context of RLS assessment within the operative segment, a discernible elevation was noted in comparison with the preoperative baseline value of 2.05 ± 0.57. Postoperative data at the 3rd month, 12th month, and the final follow-up displayed increments to 2.80 ± 0.51 (P < 0.001), 2.84 ± 0.52 (P < 0.001), and 2.83 ± 0.53 (P < 0.001), respectively, with each difference being statistically significant. These increments corresponded to rates of increase amounting to 36.59%, 38.54%, and 38.05%, sequentially. Conversely, for the non-operative segment’s RLS, no statistically significant differences were identified when comparing preoperative and postoperative data at each follow-up time point (P > 0.05). Furthermore, there were no significant differences in the RLS for the non-operative segment between preoperative and postoperative follow-up time points (F = 3.502, P = 0.0515).

Statistically significant differences in RLS were observed within the C2–3 segment (F = 269.3, P < 0.0001). The preoperative RLS was recorded as 1.84 ± 0.52, and subsequent to the surgical intervention, postoperative values demonstrated a notable increase to 2.71 ± 0.40 (P < 0.001), 2.73 ± 0.46 (P < 0.001), and 2.78 ± 0.51 (P < 0.001), as illustrated in Fig. 3.

Fig. 3

Schematic diagram of changes in the ratio of long and short diameter line (RLS) of the cervical longus at the surgical segment before and after surgery. The postoperative RLS of each surgical segment increased, which indicates that the cross-sectional morphology of the cervical longus changed from “circular” to “elliptical”

Statistically significant disparities in RLS were evident within the C3–4 segment (F = 30.93, P < 0.0001). The preoperative RLS registered at 1.87 ± 0.68, while postoperative assessments exhibited a notable escalation to 2.68 ± 0.56 (P < 0.001), 2.54 ± 0.47 (P = 0.001), and 2.73 ± 0.58 (P < 0.001), as delineated in Fig. 3.

Statistically significant discrepancies in RLS were identified within the C4–5 segment (F = 909.8, P < 0.0001). The preoperative RLS, quantified at 2.19 ± 0.56, witnessed a notable increase in the postoperative period to 2.89 ± 0.53 (P < 0.001), 3.02 ± 0.55 (P < 0.001), and 2.95 ± 0.52 (P < 0.001), as depicted in Fig. 3.

Significant statistical disparities in RLS were discerned within the C5–6 segment (F = 345.4, P < 0.0001). The initial preoperative RLS measurement, recorded at 2.18 ± 0.52, exhibited a substantial postoperative augmentation to 2.83 ± 0.54 (P < 0.001), 2.86 ± 0.56 (P < 0.001), and 2.75 ± 0.50 (P < 0.001), as illustrated in Fig. 3.

Statistically significant disparities in RLS were evident within the C6–7 segment (F = 329.5, P < 0.0001). The preoperative RLS measurement, initially quantified at 2.07 ± 0.53, demonstrated a notable postoperative increase to 2.82 ± 0.51 (P < 0.001), 2.94 ± 0.49 (P < 0.001), and 2.90 ± 0.61 (P < 0.001), as depicted in Fig. 3.

Postoperative changes of the volume of cervical longus and CESA/VBA

There were significant differences in the volume of the cervical longus between preoperative, 3 months postoperative, 12 months postoperative, and the last follow-up (F = 4109, P < 0.0001). Meanwhile, discernible alterations were observed when comparing postoperative data at the 3rd month, 12th month, and last follow-up to the preoperative baseline (8764.69 ± 492.91 mm3). Statistically significant reductions were noted in the postoperative measurements, amounting to (7762.16 ± 427.98) mm3 (P < 0.001), (7852.68 ± 422.06) mm3 (P < 0.001), and (7830.45 ± 394.44) mm3 (P < 0.001), respectively. The corresponding rates of decrease were calculated at 11.44%, 10.41%, and 10.66%, as illustrated in Fig. 4.

Fig. 4

Schematic diagram of changes in the volume of the cervical longus after surgery. The volume of the cervical longus decreased significantly at each follow-up time point after surgery, which indicates a certain degree of atrophy of the cervical longus

In the evaluation of CESA/VBA, no statistically significant differences were observed between the preoperative data and postoperative measurements at each follow-up time point (P > 0.05), as depicted in Fig. 5. However, there were significant differences in the CESA/VBA between preoperative and postoperative follow-up time points (F = 5.733, P = 0.0123).

Fig. 5

Schematic diagram of changes in the volume of cervical extensor muscles (CESA/VBA) after surgery. No statistically significant differences were identified between the preoperative CESA/VBA and the corresponding postoperative values at each follow-up time point (P > 0.05). This suggests that anterior cervical surgery may not exert a substantial impact on the cervical extensor muscles

VAS

In the evaluation of the VAS, discernible changes were noted in comparison with the preoperative baseline (3.11 ± 1.06 points). The postoperative data at the 3rd month, 12th month, and last follow-up revealed significant decreases, registering at (1.46 ± 0.92) points (P < 0.001), (1.06 ± 0.93) points (P < 0.001), and (0.82 ± 0.85) points (P < 0.001), respectively.

NDI

In the assessment of the NDI, discernible changes were observed in contrast to the preoperative baseline (16% ± 4%). The postoperative data at the 3rd month, 12th month, and last follow-up indicated statistically significant reductions, recording values of (12% ± 5%) (P < 0.001), (6% ± 4%) (P < 0.001), and (7% ± 2%) (P < 0.001), respectively.

mJOA

In relation to the mJOA scores, distinct changes were observed in comparison with the preoperative baseline (13.71 ± 1.60 points). Postoperative assessments at the 3rd month, 12th month, and last follow-up revealed statistically significant increases, registering at (15.38 ± 1.09) points (P < 0.001), (15.85 ± 0.97) points (P < 0.001), and (15.98 ± 0.82) points (P < 0.001), respectively.

Correlation analyses between the volume of cervical longus or CESA/VBA and postoperative scores (VAS, mJOA, and NDI)

Correlation analyses were separately conducted between the volume of the cervical longus or CESA/VBA and postoperative scores (VAS, mJOA, and NDI). The results revealed negative correlations between the volume of cervical longus and VAS at the 3rd (r = − 0.412, P < 0.001), 12th (r = − 0.272, P = 0.0287) month, and the last follow-up (r = − 0.391, P = 0.0013) (Fig. 6). Negative correlations were also observed between the volume of cervical longus and NDI at the 3rd (r = − 0.552, P < 0.001), 12th (r = − 0.293, P = 0.0178) month, and the last follow-up (r = − 0.459, P < 0.001) (Fig. 7). However, no significant correlation was identified between the volume of cervical longus and mJOA (P > 0.05). Furthermore, no correlation was found between CESA/VBA and postoperative scores (VAS, mJOA, and NDI) (P > 0.05).

Fig. 6

Analysis of the correlation between the volume of cervical longus and VAS score in 65 patients at the 3rd month (A), 12th month (B), and the last follow-up (C) after surgery. There was a negative correlation between the volume of cervical longus and VAS score, which indicates that the decrease in the volume of cervical longus may be the cause of postoperative neck pain

Fig. 7

Analysis of the correlation between the volume of cervical longus and NDI in 65 patients at the 3rd month (A), 12th month (B), and the last follow-up (C) after surgery. There was a negative correlation between the volume of cervical longus and NDI, which indicates that the decrease in the volume of cervical longus may be the cause of postoperative neck pain