Relationship between disability and physical activity frequency after cervical spine surgery: A linear mixed model analysis

Abstract

BACKGROUND:

There is a paucity of data on physical activity and its effects in patients after cervical spine surgery.

OBJECTIVE:

This study aimed to examine the association between physical activity and disability in patients after cervical spine surgery while also considering age, sex, pain, and central sensitization (CS)-related symptoms.

METHODS:

Participants included individuals with a cervical degenerative condition who had undergone surgery. Neck disability index, physical activity frequency, numerical rating scale for pain intensity, and short form of the CS inventory were recorded more than 1 year postoperatively. The linear mixed model was performed to examine the association between physical activity and disability.

RESULTS:

The responses of 145 participants were analyzed. The linear mixed model results showed that the stretching and light-intensity exercise frequency ($\beta =-$0.14, $p=$ 0.039) was independently associated with disability, adjusted for age, sex, pain, and CS-related symptoms. Conversely, other physical activities, such as walking and muscle strength exercises, were not associated with a disability.

CONCLUSION:

The findings emphasize the importance of performing regular physical activity, regardless of pain and CS-related symptoms.

1. Introduction

Degenerative cervical disease is a significant public health problem worldwide In individuals aged $>$ 40 years, 50%–60% have evidence of disc degeneration, 20% have foraminal stenosis and $>$ 10% have clinically significant nerve root or cord compression [1]. These percentages increase with age, and pain alone or in combination with other neurologic symptoms might require surgical intervention. Patients are typically motivated to participate in rehabilitation to maximise their outcomes postoperatively. However, there are no current standards for postoperative rehabilitation in degenerative cervical disease and no recommendations on frequency, content and rehabilitation duration [2]. Therefore, identifying the most effective strategies and postoperative rehabilitation for degenerative cervical disease has become a key public health priority.

The benefits of staying more physically active are manifold. Even later in life, regular physical activity can help older individuals prevent a decline in health-related quality of life and even improve their enjoyment of life. Physical activity has an analgesic effect that likely involves several partially overlapping mechanisms in addition to its general health benefits. Higuchi et al. [3] reported that physical activities reduce pain frequency and dysesthesia as well as a reduction of kinesiophobia, which also improves health-related quality of life in older adults who underwent lumbar spinal stenosis surgery for $>$ 1 year. Given the potential complementary effects of daily physical activities to structured exercise regimens, it is imperative to understand their impact on patients post-cervical spine surgery, a demographic where persistent pain is prevalent. However, there is currently little data on physical activity and their effects in this patient group. Given that physical activity in daily routines may have complementary effects to structured exercise bouts, identifying whether physical activity after cervical spine surgery is associated with disability and, if so, what type of physical activity, may help in planning rehabilitation programs.

There are many reasons why physical activity may be inhibited. One factor is central sensitization (CS)-induced hypersensitivity [4]. CS is defined by the International Association for the Study of Pain as “increased responsiveness of nociceptive neurons in the central nervous system to normal or subthreshold afferent input” [5]. In fibromyalgia syndrome, the representative central sensitivity syndrome, pain often worsens temporarily with increased physical activity and increased pain may interfere with this phenomenon [6]. People with chronic pain conditions, such as fibromyalgia, often show lower physical activity levels and greater sedentary behaviours than healthy control subjects [7]. Given the association between CS-related symptoms and clinical outcomes in degenerative cervical disease [8, 9], there is a need to develop prediction models that include CS-related symptoms in addition to traditional biomedical variables, including age, sex and pain to determine the relationship between physical activity frequency and disability in patients after cervical spine surgery.

Furthermore, the transition to retirement is considered a major life event that may change people’s daily routines and affect health behaviours, including daily physical activity [10]. In Japan today, the retirement age is set at 65 years [11], and physical activity levels and types might therefore differ in the two age groups ( 65 and $⩾$ 65 years). This underscores the importance of considering age-related differences when examining the association between physical activity and disability post cervical spine surgery.

This study aimed to examine the association between physical activity and disability in patients after cervical spine surgery, considering confounding factors such as age, gender, pain, and CS-related symptoms. Furthermore, as a secondary objective, it assessed how the effects of physical activity vary between elderly and non-elderly populations, considering variations across different age groups.

2. Materials and methods

2.1. Study design and ethical considerations

This cross-sectional study utilized written questionnaires mailed between August 2020 and July 2022. The study was approved by the Ethics Committee of Sapporo Maruyama Orthopedic Hospital (approval no. #000045) and was conducted in accordance with the Declaration of Helsinki. We mailed the recipients the questionnaires, research description, consent form for research participation and consent withdrawal form. Individuals who agreed to participate were asked to return the consent form with the completed questionnaire. Otherwise, a withdrawal form was submitted, and data were deleted even after completing the questionnaire. Study participants were assured that their participation would not influence their care.

2.2. Sample size

Utilizing G*Power [12], the required minimum sample size for the investigation was determined. The anticipated effect size was approximated at 0.15, with a power level of 0.80 and an alpha level of 0.05 selected. Consequently, a minimum of 127 participants was essential for employing the 12 independent variables in the regression analysis.

2.3. Participants

Participants were eligible for the study if they had a cervical degenerative condition and had undergone one of the specific types of cervical spine surgeries – either laminoplasty or anterior cervical discectomy and fusion – between December 2017 and June 2021. Degenerative diseases included spondylosis with or without myelopathy, disc herniation and ossification. To ensure a comprehensive representation, all patients who received surgery during this period were invited to participate in the study. The selection process involved reviewing medical records to identify eligible patients based on the inclusion and exclusion criteria. Participants were excluded from the study if they had had a previous fracture or luxation of the cervical vertebrae, spine, tumour, re-operation, previous surgery of the cervical spine, or diagnosed mental disease or neurologic disorder (i.e., multiple sclerosis, cerebrovascular accident). In total, 293 individuals who met the inclusion criteria and did not meet the exclusion criteria were selected. The questionnaire was mailed to all participants in July 2022, i.e., at least one year after surgery. Thirteen participants (4.4%) had unknown addresses. Consequently, we were able to mail the questionnaire to 280 individuals.

2.4. Measures

2.4.1. Clinical demographic information

Age, sex, diagnosis, date of surgery and surgical procedure were collected from the medical records. The number of postoperative days was defined as the number of days between the date of the surgery and the postmark on the return envelope.

2.4.2. Physical activities

In our study, participants were assessed on their frequency of involvement in five key types of physical activities, namely Walking, Stretching and Light-Intensity Exercises (S/LIE), Muscle Strength Exercises (MSE), Maintenance Tasks of the House and Garden (including Kitchen Garden [MTH/G]), and Paid Work. This evaluation was executed using a 5-point Likert scale, where 1 signified ‘never’, 2 indicated ‘irregularly’, 3 represented ‘once or twice a week’, 4 corresponded to ‘three or four times a week’, and 5 denoted ‘five or more days a week’. This approach was designed to capture a broad spectrum of activities that are commonly engaged in by patients post-spine surgery. The selection of this specific assessment method was underpinned by its demonstrated validity in previous studies, which have shown its effectiveness and relevance in capturing a comprehensive range of activities significant to the lifestyles of postoperative patients [3, 13].

2.4.3. Pain intensity

Pain intensity for the neck and upper limbs was measured using the 11-point Numerical Rating Scale (NRS), with 0 indicating no pain and 10 indicating the worst imaginable pain. The participants reported the average intensity of their pain over the previous month. The validity of NRS as a scale for assessing intensity of pain has already been confirmed [14].

2.4.4. Disability

The assessment of disability was conducted using the culturally adapted and validated Japanese version of the Neck Disability Index (NDI) [15], which includes 10 items scored from 0 to 5, in which a higher total score indicates greater disability. The NDI is now widely used to evaluate the efficacy of cervical surgery [16]. Previous studies have demonstrated the NDI Japanese version’s reliability and high internal consistency (Cronbach’s alpha of all items $=$ 0.89) [15].

2.4.5. Central sensitization-related symptoms

The assessment of CS-related symptoms was conducted using the culturally adapted and validated Japanese version of the Central Sensitization Inventory (CSI-9) [17]. The CSI-9 consists of nine items of CS-related symptoms, and each item is evaluated between 0 (none) and 4 (always), with a total score in the range of 0 to 36. In the Japanese version, CS-related symptoms are classified as follows: (1) emotional distress; (2) urologic and general symptoms; (3) muscle symptoms; (4) headache and jaw symptoms; and (5) sleep disturbance [17]. Previous studies have demonstrated reliability and good internal consistency validity (Cronbach’s alpha of all items $=$ 0.80) [17].

2.5. Statistical analyses

Missing values were imputed using the hot deck method. Descriptive analyses were used to evaluate the participants’ characteristics. Continuous data were expressed as the mean (standard deviation), and categorical data were expressed as counts (percentages). In our analysis, we used a linear mixed-effects model to explore the impact of physical activity on disability outcomes post-cervical spine surgery. Several factors were considered including age, sex, surgical method, days after surgery, neck pain, upper limbs pain and CS-related symptoms as covariates to limit the potential confounding factors. Before finalizing our model structure, we conducted a preliminary analysis to determine the appropriateness of incorporating mixed effects based on age groups. This decision-making process involved calculating the Intraclass Correlation Coefficient (ICC) specifically for each variable related to physical activity, assessing the variability in disability scores attributable to age group categorization (older adults $⩾$ 65 years and non-older adults 65 years). The ICC exceeding the threshold of 0.1 was deemed to justify the inclusion of mixed effects [18]. Building on this preliminary analysis, to account for potential differences in the effect of physical activity on disability after cervical spine surgery between the age group, we introduced mixed effects in the model intercepts. Specifically, we used a random intercept model where the age group (older adults $⩾$ 65 years and non-older adults 65 years) was treated as a random effect. This approach allowed us to assess whether the baseline level of disability varied across these age groups while controlling for other covariates. R ver. 4.1.2 was used for all statistical analyses, and the level of statistical significance was set to 0.05.

3. Results

3.1. Participant characteristics and data completion

A set of questionnaires was mailed to 280 participants, and responses were obtained from 169 participants (60.4%). Of these, 24 (22 who did not answer most of the questions, one with surgery for rotator cuff tears, one with surgery for osteoarthritis of the ankle) were excluded, resulting in a total of 145 (51.8%) participants who were included in the analysis (mean age: 64.4 $\pm$ 12.3 years; 99 males, 46 females; 780.1 $\pm$ 265.4 days after surgery; 69 for laminoplasty, 76 for anterior cervical discectomy and fusion) (Fig. 1). The characteristics of the study sample are summarised in Table 1.

Figure 1.

Flowchart of the participants.

Table 1.

Characteristics of the study population ($n=$ 145)

Variable
Sex ($n$) (%)
Male	99 (68.3)
Female	46 (31.7)
Age (years), mean (SD)	64.4 (12.3)
Diagnosis ($n$) (%)
Cervical spondylotic myelopathy	72 (49.7)
Cervical spondylotic radiculopathy	42 (28.9)
Cervical disc herniation	12 (8.3)
Ossification of the posterior longitudinal ligamentum	19 (13.1)
Surgery ($n$) (%)
Laminoplasty	69 (47.6)
Anterior cervical discectomy and fusion	76 (52.4)
Days after surgery, mean (SD)	780.1 (265.4)
NDI (0–50), mean (SD)	19.3 (15.5)
Physical Activities (0–5), mean (SD)
Walking	2.2 (1.4)
S/LIE	2.4 (1.4)
MSE	1.6 (1.0)
MTH/G	1.9 (1.2)
Paid works	3.0 (1.9)
NRS for neck pain (0–10), mean (SD)	3.9 (3.9)
NRS for upper limb pain (0–10), mean (SD)	3.6 (3.4)
CSI-9 (0–36), mean (SD)	14.9 (8.0)

Notes: S/LIE, stretching and light-intensity exercises; MSE, muscle strength exercises; MTH/G, maintenance tasks of the house and garden, including kitchen garden; NRS, Numerical Rating Scale; NDI, Neck Disability Index; CSI9, Short Form of Central Sensitization Inventory.

3.2. Completing missing values and descriptive statistics of the survey items

There were 27 items per respondent in this study (Physical activities: 5 items, NRS: 2 items, NDI: 10 items, CSI-9: 9 items); hence, the total number of items answered by all 145 respondents was 3770. Because 60/3770 items (1.6%) were missing, they were supplemented by the hot deck method.

3.3. Relationship between disability and physical activity after cervical spine surgery with the linear mixed model

The ICC was calculated to quantify the amount of variability in disability outcomes attributable to differences between age-based groups, yielding values of ‘NDI’ at $-$0.014 ($p=$ 0.857), ‘Paid works’ at 0.459 ( 0.001), ‘S/LIE’ at $-$0.003 ($p=$ 0.383), ‘Walking’ at 0.013 ($p=$ 0.166), ‘MSE’ at $-$0.008 ($p=$ 0.521), and ‘MTH/G’ at 0.027 ($p=$ 0.084). With the ICC for ’Paid works’ surpassing the threshold of 0.1, the introduction of mixed effects was justified and thus incorporated into our model’s intercepts. This was done to account for potential differences in the impact of physical activity on disability outcomes between older adults ($⩾$ 65 years) and non-older adults ( 65 years). However, the inclusion of mixed effects did not significantly change the model’s estimates of the baseline level of disability influenced by physical activity ($p=$ 0.247), indicating that the effects of physical activity on disability outcomes were consistent across these age groups.

Table 2 presents the linear mixed model’s results on the relationship between disability and physical activity after cervical spine surgery, Fig. 2 specifically shows a scatter plot of the relationship between S/LIE and disability. Constructed model for disability identification, S/LIE in physical activities (partial regression coefficient [B] $=-$1.53, standard error [SE] $=$ 0.73, standardised partial regression coefficient [$\beta$] $=-$0.14, $p=$ 0.039) was negatively associated with disability after cervical spine surgery. Specifically, for every increase in frequency of S/LIE, there was a decrease in disability scores. Neck pain (B $=$ 1.53, SE $=$ 0.29, $\beta =$ 0.38, $p=$ 0.001) and CS-related symptoms (B $=$ 0.86, SE $=$ 0.13, $\beta =$ 0.45, $p=$ 0.001) were positively associated with disability. Conversely, other physical activities (walking, MSE, MTH/G and paid works) were not associated with a disability after cervical spine surgery. There were no concerns with multicollinearity (variance inflation factor 2.0).

Table 2.

Result of the linear mixed model on disability after cervical spine surgery

	B	SE	$t$ values	$p$ values	$\beta$
Intercept	$-$12.64	9.90	$-$1.277	0.247	0.00
Sex	$-$1.22	2.08	$-$0.588	0.557	$-$0.04
Age	0.21	0.11	1.883	0.138	0.17
Surgical method	$-$1.34	1.99	$-$0.672	0.503	$-$0.04
Days after surgery	0.00	0.00	$-$0.433	0.666	$-$0.03
Neck pain	1.53	0.29	5.230	0.001**	0.38
Upper limb pain	0.49	0.32	1.515	0.132	0.11
CSI-9	0.86	0.13	6.513	0.001**	0.45
Physical activities
Walking	0.22	0.82	0.264	0.792	0.01
S/LIE	$-$1.53	0.73	$-$2.083	0.039*	$-$0.14
MSE	1.58	1.03	1.534	0.127	0.10
MTH/G	$-$1.05	0.80	$-$1.300	0.196	$-$0.08
Paid works	0.65	0.60	1.083	0.281	0.08

Notes: ^**p < 0.01; ^*p < 0.05. B, partial regression coefficient; SE, standard error; $\beta$, standardised partial regression coefficient; CSI-9, Short Form of Central Sensitization Inventory; S/LIE, stretching and light-intensity exercises; MSE, muscle strength exercises; MTH/G, maintenance tasks of the house and garden, including kitchen garden.

Figure 2.

Scatter plots illustrating the relationship between physical activity frequency and disability across age groups. Abbreviations: NDI, Neck Disability Index; S/LIE, stretching and light-intensity exercises.

4. Discussion

This study aimed to explore the association between physical activity frequency and disability in patients post-cervical spine surgery, utilizing a linear mixed model that incorporated variables such as age, sex, surgical method, days after surgery, pain, and CS-related symptoms. Our findings suggest that physical activity frequency was negatively associated with disability in patients after cervical spine surgery. This matches the observation in a previous study in which an association between physical activity and disability was observed [19].

A pivotal discovery in our study was the independence of the relationship between physical activity frequency and disability from pain and CS-related symptoms. This contrasts with the traditional view, often supported in literature, that pain and discomfort significantly hinder physical activity post-surgery [20]. Our findings align with the principles of non-pharmacologic chronic pain treatment, which advocate maintaining physical activity in the presence of pain [21, 22]. This suggests a paradigm shift in postoperative rehabilitation strategies, highlighting the need to design programs that encourage increased physical activity frequency for cervical spine surgery patients, irrespective of their pain or CS-related symptomatology. However, more than one year after cervical spine surgery, many patients report that they only perform physical activity approximately irregularly to once or twice a week. This may be lower than the WHO recommendation of $⩾$ 3 days a week of physical activities [23]. Therefore, additional support with a broad focus on motivational techniques and self-management support is needed to help individuals with cervical spine surgery maintain higher activity levels in the long term.

Our findings suggest that among the various types of physical activities evaluated, only the frequency of S/LIE showed a significant association with disability. This revelation is of importance, especially considering that S/LIE, characterized by its low- load nature, is advocated for patients in their post-operative phase, attributing to its potential recuperative advantages [24]. Future research should delve deeper into the mechanisms underlying these benefits and explore the potential of physical activities as a primary recommendation for post-surgical rehabilitation.

The investigation of differences in the effects of physical activity in older and non-elderly populations was a secondary aim of the study. Our findings underscored that only S/LIE exhibited a correlation with post-operative disability in patients who underwent cervical spine surgery, irrespective of their age bracket. However, preferred S/LIE may differ between the older adult and non-older adult populations. For instance, it has been reported that older Japanese people often perform traditional S/LYE, such as Japanese calisthenics and radio exercise, while middle-aged and older Japanese people often perform S/LYE, such as Pilates, yoga, and aerobics [25, 26]. Therefore, finding a S/LIE outside the care unit that suits the individual based on their specific needs and problems is important in maintaining high physical activity adherence in patients after cervical spine surgery and making physical activity part of their self-care/wellness routine [27].

The observed standardized partial regression coefficient for S/LIE of $-$0.14, while modest, significantly indicates a negative association with disability. This relationship, albeit weaker compared to factors like neck pain and central sensitization (CS)-related symptoms, is clinically significant. This significance is particularly pertinent when considering that modifying these other predictors can be challenging, as direct interventions for pain and CS-related symptoms are often complex and multifaceted [28]. The modifiable nature of physical activity, especially S/LIE, opens a practical and accessible avenue for intervention in postoperative care. Strategically incorporating S/LIE early in the rehabilitation process and progressively increasing its frequency could provide an adaptable approach, catering to patient limitations and optimizing recovery potential.

4.1. Limitations

Study results should be interpreted in light of several limitations. First, the assessment of physical activity was based on self-reports using a specialized tool, which employs a 5-point Likert scale designed to identify specific types of physical activities associated with disability in post-cervical spine surgery patients. This method was chosen for its unique ability to capture the nuances of activity types significant to our Japanese patient cohort. However, it represents a deviation from more commonly used and internationally recognized instruments such as accelerometers and the International Physical Activity Questionnaire (IPAQ). Compared to tools like the IPAQ, the limited international use and validation of our chosen tool restrict its validity and comparability on a broader scale. Second, psychological factors, such as kinesiophobia, pain catastrophizing and low self-efficacy are known barriers to staying active [29], but this study’s research has not identified the impact of these psychological factors. Future studies have been suggested to assess psychosocial factors as previous studies have supported that screening for psychosocial factors may increase the likelihood of more effective postoperative rehabilitation interventions [30]. Finally, because this was a cross-sectional study, the temporal relationship between performing regular physical activity and disability after cervical spine surgery is unclear. Therefore, future studies are recommended that would replicate our model within a prospective design.

5. Conclusions

Our study demonstrated a negative association between physical activity frequency and disability in patients post-cervical spine surgery, independent of pain and CS-related symptoms. Notably, this beneficial effect of physical activity on disability was consistent across various age groups, both below and above 65 years. These findings highlight the critical importance of incorporating regular physical activity into the rehabilitation process for post-surgical patients, regardless of their age or symptomatology.

Author contributions

YK and DH conceived the idea of the study. DH developed the statistical analysis plan and conducted statistical analyses. TM and YW contributed to the interpretation of the results. YK drafted the original manuscript. DH and TT supervised the conduct of this study. All authors reviewed the manuscript draft and revised it critically on intellectual content. All authors approved the final version of the manuscript to be published.

Data availability

The datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request.

Funding

His study did not receive funding in any form.

Ethical approval

The study was approved by the ethics committee of Sapporo Maruyama Orthopedic Hospital (approval no. #000045). Written informed consent was obtained from the patients for publication of this study.

Conflict of interest

The authors declare that they have no conflicts of interest regarding the publication of this paper.