The Longitudinal Changes of Cervical Sagittal Alignment in a Non-surgical Cohort: A 12-Year Longitudinal Study

Abstract

Study Design

A retrospective longitudinal observational cohort study.

Objective

This study aimed to quantitatively evaluate the natural progression of cervical sagittal alignment parameters in non-surgical patients over a mean 12-year follow-up period, with a specific focus on age- and sex-related variations.

Methods

Radiographic data were analyzed from 198 non-surgical patients (mean age: 50.6 years) over a mean follow-up duration of 142 months (range: 96–203 months). Key parameters measured included C2-7 Cobb angle, range of motion (ROM), cervical sagittal vertical axis (SVA), T1 slope, C7 slope, thoracic inlet angle (TIA), and neck tilt. Longitudinal changes were evaluated alongside the influences of age and sex.

Results

The cohort exhibited significant, age-accelerated progression in cervical sagittal alignment. Overall cervical ROM decreased markedly from 50.10° to 41.14° (P<0.001). Concurrently, a progressive trend toward anterior sagittal imbalance was observed, characterized by significant increases in C7 slope, T1 slope (24.86° to 26.41°), TIA, and SVA (22.38 mm to 24.59 mm). In contrast, the C2-7 Cobb angle and neck tilt remained stable. Advanced age was a strong independent predictor of deterioration. Furthermore, males experienced significantly greater reductions in ROM and more pronounced increases in SVA compared to females.

Conclusion

The natural aging process of the cervical spine is characterized by progressive stiffening and a trajectory toward anterior sagittal malalignment, primarily mediated through a compensatory increase in T1 slope that is not matched by a reciprocal increase in lordosis. These data highlight the need to account for natural physiological drift when evaluating ultra-long-term iatrogenic surgical effects.

Introduction

Cervical spinal degeneration is an inevitable age-related process characterized by reduced disc height, osteophytes, and instability.^1,2 Smith et al. ³ reported that the prevalence of asymptomatic cervical spinal cord compression increases significantly with age, providing the anatomical substrate for the development of pathological conditions such as cervical spondylotic myelopathy or cervical radiculopathy. ⁴ With the accelerating global trend of population aging, elucidating the physiological aging process preceding pathological compression has emerged as a major focus of current research.

Beyond localized degenerative features, the sagittal alignment parameters of the cervical spine are gaining increasing attention in the scientific community. ⁵ Parameters such as the C2–C7 lordosis angle, T1 slope, and cervical sagittal vertical axis are essential for maintaining a favorable biomechanical environment of the spine.^6-8 Although numerous studies have established normative ranges for these parameters across different ages and sexes, and have identified positive correlations between advancing age and increased values of parameters like C7 slope and T1 slope,^9,10 the majority of this evidence derives from cross-sectional studies. A fundamental limitation of such designs is their inability to accurately capture the true temporal trajectory of degeneration at the individual level; they cannot distinguish whether observed differences reflect intergenerational disparities or genuine intra-individual structural changes over time.^2,7,11 Consequently, the actual rate of structural change during cervical degeneration, as well as the precise effects of age and sex, remain incompletely understood.

Longitudinal data on cervical sagittal parameters in clinical cohorts remain scarce, particularly studies exceeding 96 months that incorporate baseline ROM and TIA. To address this gap, we retrospectively analyzed 198 volunteers with a mean follow-up of 10 years, testing the central hypothesis that age governs the velocity of degeneration while sex influences its phenotypic pattern. Our specific objectives were to: (1) quantify the trajectory of cervical sagittal alignment changes; (2) assess the distinct influences of age and sex; and (3) establish baseline data to distinguish physiological aging from iatrogenic outcomes.

Materials and Methods

Study Design and Population

This study utilizes a retrospective diagnostic effectiveness design approved by the ethics review board of Peking University Third Hospital (M2022408). Informed consent was obtained from all individual participants included in the study. Our research included participants who visited Peking University Third Hospital from January 2006 to January 2015 and had complete lateral and hyperextension or hyperflexion X-rays of the cervical vertebrae. The study included participants who met the following criteria: (1) aged between 18 and 80 years, (2) second visit at least 96 months away from the first visit, (3) had complete X-ray results of the lateral and hyperextension and hyperflexion positions of the cervical vertebrae. Participants who met one of the following criteria were excluded: (1) underwent any spinal surgery prior to the second visit, (2) diagnosed with any spinal trauma or deformity, (3) diagnosed with spinal tumors, infections, or other related pathologies, (4) diagnosed with musculoskeletal disorders such as rheumatoid arthritis or ankylosing spondylitis.

Image Data Acquisition and Radiographic Measurement

Participants underwent plain radiography of the cervical spine in neutral lateral, as well as maximum flexion and extension positions, with a standard tube-to-film distance of 1.5 meters. All radiographic data were stored in a Picture Archiving and Communication System (PACS; General Electric Company, Boston, MA). Radiographic evaluations were performed at two distinct time points separated by a minimum interval of 96 months.

To comprehensively assess cervical alignment, the following parameters were systematically measured:

• (1) C2–C7 Cobb angle: the angle formed between lines parallel to the lower endplates of C2 and C7;

• (2) C2-7 Extension (C2-7 Ex) and Flexion (C2-7 Fle) angles: the C2–C7 Cobb angles measured in maximum extension (recorded as a positive value) and maximum flexion (recorded as a negative value), respectively;

• (3) Subaxial cervical range of motion (C2-C7 ROM): the absolute value of the difference between the C2-7 Ex and C2-7 Fle angles;

• (4) Cervical sagittal vertical axis (SVA): the horizontal distance from the posterosuperior corner of the C7 vertebral body to a plumb line dropped from the center of the C2 vertebral body;

• (5) T1 slope (T1S): the angle between the horizontal plane and the upper endplate of T1;

• (6) C7 slope (C7S): the angle between the horizontal plane and the lower endplate of C7;

• (7) Thoracic inlet angle (TIA): the angle formed by the intersection of a line connecting the center of the T1 upper endplate to the upper end of the sternum and a line perpendicular to the T1 upper endplate;

• (8) Neck tilt (NT): the angle between a vertical line originating from the upper end of the sternum and a line connecting the upper end of the sternum to the center of the T1 upper endplate;

• (9) Spino-cranial angle (SCA): the angle between the C7 slope line and a straight line joining the center of the C7 lower endplate to the center of the sella turcica.

Overall cervical alignment was defined as kyphotic when the neutral C2–C7 Cobb angle was negative, and lordotic when positive (Figure 1A–C).

Figure 1.

Sagittal parameters of cervical X-rays. In lateral cervical X-ray (A), a represents the Cobb angle, d represents the SVA, e represents T1S, f represents TIA, g represents NT, and h represents SCA. In hyperextension lateral cervical X-ray (B), b represents C2-7Ex. In hyperflexion lateral cervical X-ray (C), c represents C2-7Fle

Reliability and Measurement Error Analysis

To ensure measurement accuracy and quantify precision, a formal reliability analysis was conducted. All radiographic parameters were measured independently by three spine surgeons adhering to a predefined protocol. Intra-observer reliability was assessed by having one observer re-measure a random subset of 50 radiographs after a minimum 2-week interval, blinded to initial results. Inter-observer reliability was evaluated using the initial measurements from all three observers on the same subset. Intraclass correlation coefficients (ICC) were calculated using a two-way random-effects model (absolute agreement) for inter-observer reliability and a two-way mixed-effects model (absolute agreement) for intra-observer reliability. ICC values greater than 0.75 were considered excellent. The standard error of measurement (SEM) and the minimal detectable change at the 95% confidence level (MDC₉₅) were derived to distinguish true change from measurement error.

Statistical Analysis

Statistical analyses were performed using SPSS 26.0 (SPSS, Chicago, IL, USA). The Shapiro–Wilk test was applied to assess the normality of the data. Continuous variables are presented as mean ± standard deviation, and categorical variables as frequency (percentage). Comparisons of categorical variables were performed using Pearson’s chi-square test or Fisher’s exact test, as appropriate. Continuous variables with normal and non-normal distributions were compared between subgroups using independent Student’s t-tests and Mann–Whitney U tests, respectively. Differences in continuous variables between the two follow-up time points were evaluated with paired t-tests or Wilcoxon signed-rank tests, depending on distribution. One-way ANOVA was used for comparisons among more than two groups. A P value <0.05 was considered statistically significant. The association between two variables was assessed by Pearson correlation analysis, with statistical significance set at P < 0.001.

Results

Demographic Data of Study Population

A total of 198 non-surgical patients were included in this retrospective longitudinal study, consisting of 63 males and 135 females. The mean age at the initial visit was 50.59 ± 12.37 years (range, 19–79 years). For descriptive purposes, the cohort was initially stratified by age into three groups: Group 1 (<40 years, n=40), Group 2 (40–60 years, n=113), and Group 3 (>60 years, n=45). However, recognizing the unbalanced sample sizes and that chronological age is intrinsically a continuous linear variable, our primary conclusions regarding age-related progression rely on the continuous linear regression models rather than inter-group comparisons. The mean follow-up duration was 142.36 ± 26.28 months (range, 96–203 months). There were no statistically significant differences in follow-up duration between age groups or genders (P>0.05).

Longitudinal Changes in Cervical Alignment and Motion

Over a minimum follow-up of 8 years (mean 11.9 years), significant alterations in cervical biomechanics were observed in the overall cohort (Table 1). The cervical range of motion (ROM) significantly decreased from 50.10° ± 17.11° to 41.14° ± 16.03° (P<0.001), driven by significant reductions in both extension (P=0.023) and flexion (P<0.001) angles.

Table 1.

Comparison of Radiographic Parameters Between Initial and Final Follow-Ups in the Overall Cohort

​	Initial follow-up	Final follow-up	p-value
Cobb	10.22±12.17	11.14±11.95	0.202
C2-7 Ex	29.05±13.40	26.78±12.88	0.023*
C2-7 Fle	-21.05±11.15	-14.37±12.14	<0.001*
ROM	50.10±17.11	41.14±16.03	<0.001*
C7S	20.36±7.84	22.36±8.99	0.003*
T1S	24.86±7.90	26.41±8.77	<0.001*
SVA	22.38±9.92	24.59±12.22	0.004*
TIA	74.54±10.46	76.95±11.14	<0.001*
NT	49.73±7.79	50.41±7.76	0.157
SCA	78.06±9.61	77.71±8.89	0.549
T1S-Cobb	14.62±8.59	15.30±9.57	0.490

*p<0.05.

Regarding sagittal alignment, the cohort exhibited a tendency toward an anterior imbalance over time. Significant increases were noted in C7 slope (C7S; 20.36° vs. 22.36°, P=0.003), T1 slope (T1S; 24.86° vs. 26.41°, P<0.001), thoracic inlet angle (TIA; 74.54° vs. 76.95°, P<0.001), and C2–7 sagittal vertical axis (SVA; 22.38 mm vs. 24.59 mm, P=0.004). Notably, the C2–7 Cobb angle, neck tilt (NT), and spino-cranial angle (SCA) remained stable during the follow-up period (P>0.05).

Impact of Age on Sagittal Alignment

Difference Analysis

Age-related stratification was evident at both the initial and final assessments (Table 2). Older participants (Group 3) consistently demonstrated significantly lower ROM and larger sagittal alignment parameters (C7S, T1S, TIA, and NT) compared to younger cohorts (Group 1,2) at both time points (P<0.05 for inter-group comparisons). While Cobb angle was significantly larger in the older group (Group 3) at the final follow-up, SVA showed significant inter-group differences primarily at the final follow-up (0.039).

Table 2.

Comparison of Radiographic Parameters of Different Groups of Age Between Initial and Final Follow-Ups

​	Group 1		Group 2		Group 3		P1	P2
	Initial follow-up	Final follow-up	Initial follow-up	Final follow-up	Initial follow-up	Final follow-up
Cobb	8.21±13.45	8.57±11.18	9.89±11.72	9.92±10.81	12.93±11.80	16.58±13.72	0.182	0.002*
C2-7 Ex	31.31±13.83	28.20±15.03	29.26±12.93	25.87±11.74	26.49±4.07	27.50±13.35	0.247	0.489
C2-7 Fle	-25.75±12.36	-17.84±12.35	-20.88±10.09	-15.91±10.04	-17.32±1.28	-7.42±14.22	0.002*	<0.001*
ROM	57.06±18.22	46.36±17.80	50.14±16.98	41.78±14.29	43.81±14.07	34.92±16.88	0.002*	0.003*
C7S	19.15±7.17	19.20±6.80	20.31±7.75	21.23±8.03	23.32±8.18	28.06±10.57	0.033*	<0.001*
T1S	24.27±7.15	23.84±7.18	24.29±8.02	25.28±7.77	26.84±8.07	31.54±10.46	0.163	0.001*
SVA	22.48±15.03	22.01±9.86	21.53±9.60	23.58±11.13	23.87±11.65	28.89±15.45	0.428	0.039*
TIA	68.92±9.67	70.82±9.99	75.21±10.25	77.31±11.21	77.84±9.91	81.49±9.60	<0.001*	<0.001*
NT	44.76±7.63	47.03±7.87	50.63±8.26	51.45±8.64	50.94±7.62	49.82±6.99	<0.001*	0.003*
SCA	80.97±9.73	80.22±7.44	77.58±9.84	78.35±9.11	76.68±8.54	73.88±8.49	0.087	0.002*
T1S-Cobb	16.06±9.88	15.28±9.38	14.39±8.10	15.36±8.58	13.90±8.63	14.96±12.04	0.472	0.557

*p<0.05.

P1 P-value of the different between three groups at the initial follow-up, P2 P-value of the different between three groups at the final follow-up.

Correlations of Parameters

The rate of sagittal deterioration was significantly correlated with age (Table 3). Pearson correlation analysis demonstrated positive correlations between age at enrollment and the longitudinal increase (Δ) in SVA (r=0.202, P<0.001), C7S (r=0.281, P<0.001), T1S (r=0.291, P<0.001) and TIA (r=0.168, P<0.001).

Table 3.

Correlation Between Age and Longitudinal Changes in Radiographic Parameters

​	Age	Cobb	C2-7 Ex	C2-7 Fle	ROM	SVA	C7S	T1S	TIA	NT	SCA	T1-Cobb
Age	​	0.144	-0.118	0.057	-0.055	0.202*	0.281*	0.291*	0.168*	-0.077	0.122	0.072
Cobb	​	​	-0.304*	0.223	-0.097	-0.268*	0.556*	0.556*	0.372*	-0.093	0.722*	-0.722*
C2-7 Ex	​	​	​	0.097	0.810*	0.013	-0.221	0.237	-0.128	0.071	-0.155	0.163
C2-7 Fle	​	​	​	​	0.663*	-0.062	0.153	0.120	0.090	-0.016	0.172	-0.165
ROM	​	​	​	​	​	-0.026	-0.074	-0.108	-0.043	0.044	-0.016	0.025
SVA	​	​	​	​	​	​	0.474*	0.449*	0.235*	-0.163	-0.257*	0.691*
C7S	​	​	​	​	​	​	​	0.817*	0.511*	-0.176	0.553*	0.040
T1S	​	​	​	​	​	​	​	​	0.665*	-0.161	0.427*	0.173
TIA	​	​	​	​	​	​	​	​	​	0.603*	0.305*	0.112
NT	​	​	​	​	​	​	​	​	​	​	0.009	-0.024
SCA	​	​	​	​	​	​	​	​	​	​	​	-0.501

*p<0.001.

Multiple Regression Analysis and Linear Regression Model

Linear regression analysis demonstrated a statistically significant relationship between advanced age and the progression of anterior imbalance (Figure 2). However, the extremely low R2 values (ranging from 0.028 to 0.085) indicate that chronological age accounts for only a marginal fraction (less than 9%) of the overall variance in sagittal deterioration. This suggests that while age is a contributing factor, it is a weak independent predictor. The regression equations for the change in parameters are as follows:

$$\mathrm{\Delta }\text{SVA}=-4.873+0.140\times \text{Age}\hspace{0.17em}({\mathrm{R}}^2=0.041,\mathrm{P}<0.001)$$

$$\mathrm{\Delta }\mathrm{C}7\mathrm{S}=-5.464+0.140\times \text{Age}\hspace{0.17em}\left({\mathrm{R}}^2=0.079.\hspace{0.17em}\mathrm{P}<0.001\right)$$

$$\mathrm{\Delta }\mathrm{T}1\mathrm{S}=-6.832+0.166\times \text{Age}\hspace{0.17em}\left({\mathrm{R}}^2=0.085,\mathrm{P}<0.001\right)$$

$$\mathrm{\Delta }\text{TIA}=-3.669+0.120\times \text{Age}\hspace{0.17em}\left({\mathrm{R}}^2=0.028,\mathrm{P}<0.001\right)$$

Figure 2.

Scatter plots with linear regression lines illustrating the relationship between age at enrollment and the longitudinal changes (Δ) in SVA (A), C7S (B), T1S (C), and TIA (D) over the follow-up period

No significant correlation was found between age and the longitudinal changes in Cobb angle, ROM, NT, or SCA.

Impact of Gender on Sagittal Alignment

Difference Analysis

Sexual dimorphism was stable throughout the study period (Table 4). Males consistently demonstrated significantly larger cervical parameters (Cobb, C7S, T1S, SVA, TIA, and NT) than females at both the initial and final assessments (P<0.05).

Table 4.

Comparison of Radiographic Parameters of Different Groups of Gender Between Initial and Final Follow-Ups

​	Male		Female		P1	P2
	Initial follow-up	Final follow-up	Initial follow-up	Final follow-up
Cobb	13.65±13.36	14.70±13.84	8.66±11.24	9.51±10.57	0.011*	0.010*
C2-7 Ex	30.91±12.92	26.29±3.78	28.18±13.58	27.00±12.46	0.175	0.729
C2-7 Fle	- 19.80±10.67	- 11.97±13.84	- 21.64±11.36	-15.49±11.15	0.271	0.139
ROM	50.71±14.34	38.27±16.61	49.82±18.30	42.49±15.64	0.709	0.092*
C7S	22.79±7.75	25.06±9.12	19.84±7.73	21.11±8.68	0.015*	0.005*
T1S	27.14±8.05	28.78±9.33	23.80±7.62	25.31±8.31	0.007*	0.009*
SVA	23.75±11.23	27.87±12.89	21.74±9.22	23.06±11.63	0.130	0.008*
TIA	77.75±9.40	81.05±10.29	73.04±10.62	75.03±11.04	0.002*	<0.001*
NT	50.61±7.36	52.18±7.28	49.32±7.97	49.48±7.85	0.266	0.024*
SCA	75.23±9.42	74.58±8.93	79.38±9.44	79.17±8.52	0.005*	0.001*
T1S-Cobb	13.49±10.04	14.08±10.15	15.14±7.80	15.80±9.28	0.251	0.156

*p<0.05.

P1 P-value of the different between male and female at the initial follow-up, P2 P-value of the different between male and female at the final follow-up.

Correlations of Parameters

Gender significantly influenced the magnitude of longitudinal changes for specific parameters (Table 5). Males experienced a significantly greater loss of ROM compared to females (ΔROM: -12.45° ± 15.11° vs. -7.33° ± 18.95°, P=0.043). Furthermore, the anterior shift of the cervical spine was more pronounced in males, as evidenced by a significantly larger increase in SVA (ΔSVA: 4.12 ± 9.78 mm vs. 1.31 ± 7.85 mm, P=0.048). No significant gender differences were observed in the longitudinal progression of Cobb angle, C7S, T1S, or TIA.

Table 5.

Correlation Between Gender and Longitudinal Changes in Radiographic Parameters

​	Male	Female	p-value
Cobb	1.05±11.18	0.85±9.49	0.904
C2-7 Ex	4.62±11.68	1.18±14.18	0.074
C2-7 Fle	7.83±10.22	6.15±10.73	0.292
ROM	12.45±15.11	7.33±18.95	0.043*
C7S	2.40±6.07	1.27±6.09	0.228
T1S	1.64±7.53	1.51±6.83	0.905
SVA	4.12±9.78	1.31±7.85	0.048*
TIA	3.30±8.55	2.00±9.00	0.328
NT	1.57±5.71	0.26±7.04	0.166
SCA	0.65±7.52	0.21±8.53	0.878
T1S-Cobb	0.59±8.62	0.65±8.42	0.962

*p<0.05.

Discussion

This study provides a longitudinal analysis of cervical sagittal alignment in a non-surgical cohort with a mean follow-up of nearly 12 years. Our findings delineate the natural trajectory of cervical aging: a progressive decline in the ROM and a tendency toward forward sagittal imbalance, characterized by significant increases in the C7S, T1S, and SVA. Notably, despite these changes in downstream parameters, the C2–C7 Cobb angle and NT remained relatively stable. Understanding these age-related natural variations is paramount for distinguishing pathological progression from physiological senescence.

Cervical degeneration is a ubiquitous physiological process, typically initiated by disc dehydration and height loss, which subsequently leads to biomechanical instability.^4,12 The significant positive correlation observed in this study between age and increased anterior cervical malalignment (elevated T1S, C7S, and SVA) aligns with previous cross-sectional findings. Chen et al. ¹⁰ noted that an increased T1S is common among elderly Asian populations, while C7S, as a surrogate for T1S, has also been reported to correlate positively with age, ¹³ Wang et al. ¹⁴ found that SVA increases with advancing age in an analysis of radiographs from 701 asymptomatic subjects. Furthermore, we found that the loss of ROM was significantly correlated with aging, supporting the “cascade” theory proposed by Kirkaldy-Willis et al. ¹⁵ This theory suggests that segmental instability induces osteophytosis and ligamentous hypertrophy to restabilize the spine, albeit at the expense of mobility.

A core finding of this study is the confirmation that cervical degeneration does not progress at a constant rate but accelerates with age, which is fully validated by long-term longitudinal data. Previous cross-sectional studies could only reveal parameter differences between different age groups and failed to reflect the temporal trajectory of intra-individual degeneration. In contrast, through a mean follow-up of 11.9 years, this study found that the rates of change in T1S, C7S, TIA, and SVA showed statistically significant, albeit weak, linear correlations with age (regression equations: R²=0.028-0.085, all P<0.001), and the magnitude of parameter changes (e.g., increase in SVA, decrease in ROM) in the elderly group (>60 years) at the end of follow-up was significantly greater than that in the younger group (<40 years). This result suggests that there is an “acceleration inflection point” in cervical degeneration. With increasing age, the cumulative effects of underlying lesions such as disc degeneration and facet joint wear gradually become apparent, leading to a significant acceleration in the rate of sagittal imbalance and mobility decline, rather than a simple linear progression.

In this accelerated degenerative process, the increase in T1S and the stability of NT are key mechanisms driving sagittal imbalance, which are closely associated with pathological changes such as osteophyte formation, ligamentous hypertrophy, and cervical muscle function decline. Existing research indicates that as humans age, the increase in lumbar kyphosis—driven by osteoporosis and vertebral wedging—necessitates a steeper T1 slope.^9,16 Knott PT et al. ⁸ highlighted the initiating role of T1S in sagittal imbalance. According to Dubousset’s “cone of economy” theory, the spine, including the cervical region, must adjust to maintain a horizontal gaze. ¹⁷ Importantly, cervical degeneration does not occur in isolation. Extensive literature confirms that the cervical, thoracic, and lumbar segments exhibit both static morphological changes and dynamic interactions.^18-20 Although our study evaluated only cervical data, recognizing this inter-segmental linkage is crucial. Cervical sagittal alignment is intimately tied to global spinal curvature, reflecting dynamic compensations from thoracic kyphosis (TK), lumbar lordosis (LL), and pelvic tilt. For instance, Lee et al. ²¹ delineated a transmission chain affecting cervical alignment: pelvic incidence (PI) → sacral slope (SS) → LL → TK → cervical spine. Similarly, Charles et al. ²² found that cervical parameters vary with PI, noting strong correlations among C2–C7 lordosis, C7/T1 slopes, and TK. Without whole-spine radiographic data, we cannot systematically analyze these exact inter-segmental effects in our cohort. Nonetheless, this global spino-pelvic interaction remains an essential context for interpreting our findings.

Therefore, in an ideal state of complete compensation, an increase in T1S should be matched by a reciprocal increase in cervical lordosis to maintain cephalic balance. ²³ However, in our cohort, while T1S increased from 24.86° to 26.41° (P<0.001), the C2–C7 Cobb angle did not increase proportionally to offset its effect, ultimately leading to forward cervical displacement (increased SVA from 22.38 mm to 24.59 mm, P=0.004). Meanwhile, as a parameter dependent on rigid osseous structures—specifically the T1 vertebral body and the manubrium sternum^24-26 —NT showed no significant intra-individual changes during the 12-year follow-up. Based on the geometric relationship TIA = T1S + NT established by Lee et al., ²⁶ the increase in T1S directly drives the elevation of TIA when physiological NT is preserved, further exacerbating sagittal imbalance. In this process, segmental instability caused by disc degeneration induces osteophyte formation and ligamentous hypertrophy. ¹⁵ Although these pathological changes can enhance spinal stability to a certain extent, they also lead to cervical stiffness and ROM reduction. Furthermore, the age-related increase in fatty infiltration of the posterior cervical extensor muscles compromises muscle strength and stability. ²⁷ With an increase in T1 slope, the lever arm of the head is elongated, which leads to an exponential rise in muscular loading. This initiates a vicious cycle characterized by muscle fatigue, atrophy and progressive sagittal malalignment. Importantly, paraspinal muscle quality is not merely a secondary consequence of spinal degeneration but a critical determinant of clinical outcomes. Recent evidence demonstrates that the degree of paraspinal fatty infiltration correlates closely with chronic pain profiles and serves as a key parameter in surgical decision-making. ²⁸ Evaluating muscle structural integrity can effectively assist clinicians in differentiating candidates for surgical intervention from those suitable for conservative management. Consequently, incorporating muscle quality assessments yields a more comprehensive understanding of cervical senescence.

It is tempting to speculate on the gender-specific patterns revealed by our data—namely, the more significant ROM loss and SVA increase in males. One plausible hypothesis is that males primarily utilize cervical stiffening to counteract forward imbalance. This would align with the framework of DCM pathophysiology, which encompasses both static compression and dynamic instability, ⁴ and might imply that males are more susceptible to the former, while females, who retain greater flexibility, could be predisposed to the latter. ⁵ This gender disparity may be partially attributed to biological factors, particularly the influence of estrogen on collagen metabolism and ligamentous laxity, which might mitigate cervical stiffening in females compared to males.^29,30 Thus, the observed progressive forward imbalance may represent an anatomical substrate for degenerative diseases; however, this mechanistic interpretation remains a hypothesis to be tested in future research that directly correlates longitudinal imaging with clinical outcomes.

As correctly noted, our cohort consists of patients presenting for clinical evaluation, not truly asymptomatic volunteers. Therefore, the temporal changes observed in our study likely represent the natural history of early/mild cervical spondylotic degeneration rather than pure physiological aging in a completely healthy population. However, this non-surgical clinical cohort is highly representative of the actual patient population that spine surgeons evaluate in daily practice, making this baseline practically relevant when comparing outcomes against surgical cohorts. While our findings provide a longitudinal reference, the extremely small magnitude of natural change precludes us from directly using this as a strict numerical baseline to subtract from surgical cohorts. Instead, it conceptually reminds clinicians that not all post-operative alignment loss is entirely iatrogenic. In the literature regarding non-fusion surgeries such as laminoplasty, postoperative loss of lordosis and ROM is frequently attributed primarily to the surgical intervention.^7,31-33 However, the lack of long-term non-surgical control groups may lead to a partial overestimation of the iatrogenic effect. Our data show that even without surgery, there is an inevitable, albeit gradual, physiological trajectory involving a loss of approximately 9° of ROM and a subtle 2.2 mm increase in SVA over 12 years of natural aging. While we acknowledge that the magnitude of these natural age-related changes (e.g., 1.5°–3° angular differences) is relatively small and may not fully account for the large, acute alignment shifts often observed postoperatively, they nonetheless represent a constant underlying background of senescence. Existing long-term follow-up data on non-fusion spinal surgeries further support this notion. For instance, DH Lee et al. ³⁴ reported that age was significantly associated with sagittal imbalance (C2–7 > 40 mm) at 2 years following cervical laminoplasty (p = 0.027). Furthermore, Mun et al. ³⁵ found that postoperative changes in C2–7 SVA after cervical laminoplasty were significantly more pronounced in elderly patients than in their younger counterparts (p < 0.01). Therefore, when evaluating the ultra-long-term (e.g., >10 years) efficacy and sagittal stability of non-fusion procedures, these gradual age-related changes should be taken into consideration. Future studies, such as those by Guo et al., ³⁶ could refine their evaluations of different surgical techniques by incorporating the natural history corrections provided here. By incorporating this longitudinal baseline of natural aging, future studies could refine their evaluations, and surgeons could more comprehensively counsel patients on long-term expectations, recognizing that postoperative alignment evolution may be a combination of both surgical impacts and natural degenerative progression.

This study has several limitations. First, the retrospective design and limited sample size may restrict the generalizability of the findings; therefore, future studies should incorporate larger cohorts and prospective methodologies to better identify potential risk factors. Second, this cohort consisted of patients who presented for cervical evaluation and were managed conservatively. They are not entirely asymptomatic, healthy volunteers, which may introduce a selection bias regarding the natural history of spinal deterioration. Third, the current assessment was confined to two-dimensional radiographic parameters. The lack of concurrent patient-reported outcomes such as the Neck Disability Index and advanced diagnostic modalities limits our ability to fully elucidate the underlying mechanisms of the degenerative cascade. Notably, the absence of magnetic resonance imaging and three-dimensional finite element analysis prevents a quantitative evaluation of soft tissue deterioration, particularly the paraspinal musculature, as well as localized biomechanical alterations including intervertebral disc pressure and facet joint stress. ³⁷ Future investigations integrating multimodal imaging and biomechanical modeling are warranted to comprehensively validate these structural and kinematic findings. Fourth, our radiographic analysis focused on the subaxial spine (C2-C7). Due to limitations in retrospectively visualizing the occipitocervical junction on some older historical radiographs, we did not evaluate Occiput-C2 (O-C2) alignment parameters (e.g., the angle subtended by McGregor’s line to the C2 inferior endplate). Since the upper cervical spine contributes significantly to total cervical ROM and lordotic compensation, the omission of O-C2 dynamics is a notable limitation that likely accounts for a portion of the unexplained variance in our models. Fifth, although age is a significant predictor of sagittal imbalance, the low R² values (0.028-0.085) indicate that chronological age explains only a limited portion of the variance. Finally, as discussed earlier, the lack of concurrent global spino-pelvic radiographic data limits our ability to quantitatively model the dynamic inter-segmental compensatory mechanisms; future prospective studies must integrate whole-spine variables to construct a more comprehensive model of cervical senescence.

ORCID iDs

Xiaoxiong Yang https://orcid.org/0000-0002-7035-191X

Feifei Zhou https://orcid.org/0000-0001-9114-9123