Associations between disc space narrowing, anterior osteophytes and disability in chronic mechanical low back pain: a cross sectional study

A descriptive cross-sectional study was conducted on consecutive patients with chronic mechanical low back pain who attended the rheumatology clinic, National Hospital of Sri Lanka, Colombo, from May 2012 to May 2014. Both male and female patients of Sri Lankan origin with chronic mechanical low back pain aged 20 to 69 years were recruited to the study. Both patients with and without x-ray evidence of LDD and spondylolisthesis were included. Low back pain was defined as pain, muscle tension, or stiffness localized below the costal margin and above the inferior gluteal folds, with or without pain radiating into the leg [19]. Back pain during day time worsening in the latter part of the day due to movements was considered to be due to a mechanical cause [28]. Chronicity was defined as pain on most days of the week for at least three months [2]. Patients with back pain due to inflammatory causes (seronegative spondyloarthropathies, diffuse idiopathic skeletal hyperostosis, rheumatoid arthritis), visceral origin (urinary tract infections, inflammatory pelvic disease), systemic infections affecting spine (spinal tuberculosis), metabolic bone diseases (osteoporosis and osteomalacia), fractures in the vertebral column, past surgeries in the spine, and spinal tumours were excluded. Pregnant females and patients who refuse to participate in the study were also excluded. The study was carried out in accordance with the Declaration of Helsinki and with the approval of the Ethics Review Committee of the Faculty of Medicine, University of Colombo. Patients who fulfilled the inclusion and exclusion criteria were recruited to the study after obtaining written informed consent.

Demographic (age and gender) and clinical data (intensity of pain, severity of disability, presence of pain radiating into legs, and BMI) were recorded using a pretested interviewer administered questionnaire and clinical examination. The intensity of pain was measured using a 101 (0 to 100) point numeric rating scale. Patients were asked to score the average intensity of pain experienced during the past 7 days out of 100 [29‐31]. Disability was assessed using the Modified Oswestry Disability Index (MODI). MODI is a low back pain specific disability questionnaire with ten items which assess pain and its impact on the activities of daily living including personal care, lifting, walking, sitting, standing, sleeping, travelling, social work, home and work duties. Each item has six responses where higher values represent greater disability. Sum of responses was calculated and presented as a percentage [32, 33]. Pain radiating into legs was positive if the pain radiated below the knee of either one or both legs. Height (cm) and weight (kg) of the patients were recorded with light clothing and without shoes to the nearest 0.1 cm and 0.1 kg, respectively, and BMI was calculated (kg/m²) [34]. International cut off values were used for categorisation of BMI [35].

Static lateral lumbar x-rays were obtained from all patients. Patients were in lateral recumbent position on the table flexing the knees and hips just enough to achieve comfortable position and a small sandbag was kept between the knees. Midaxillary plane was aligned to the middle of the table and the central x-ray beam was directed perpendicular to the body of the L3 vertebra [36]. Collected lateral lumbar x-rays were evaluated by a consultant radiologist blinded to the clinical details of the patients. The intervertebral disc spaces (L1/L2 to L5/S1) in lateral lumbar x-rays were assessed for the disc space narrowing, anterior osteophytes and lumbar spondylolisthesis. Reduction of the height of the disc space compared to the adjacent normal disc space was defined as the disc space narrowing and presence of bony outgrowths of the vertebral body arising from the borders of superior and inferior surfaces extending anteriorly was defined as anterior osteophyte. Disc space narrowing was graded as follows: grade 0 = none; grade 1 = definite (mild) narrowing; grade 2 = moderate to severe narrowing. Anterior osteophyte was graded as follows: grade 0 = none; grade 1 = small osteophyte and grade 2 = moderate to large osteophyte. Based on these features, overall grading was given for the LDD: grade 0 = normal (grade 0 disc space narrowing and grade 0 anterior osteophyte); grade 1 = grade 1 disc space narrowing and/or grade 1 anterior osteophyte; grade 2 = grade 2 disc space narrowing and/or grade 2 anterior osteophyte (Fig. 1) [37]. End plate sclerosis was not taken into account due to its low interobserver reliability [9, 15]. A particular grade of disc space narrowing/anterior osteophyte/LDD was identified for each of the lumbar levels, and the highest available grade out of the five lumbar levels was used as the final grade for that particular spine. Lumbar spondylolisthesis was defined as presence of displacement of one vertebral body relative to the next most inferior vertebral body and assessed in lateral recumbent lumbar x-ray [11]. However the ability to assess the spondylolisthesis in lateral recumbent lumbar x-ray is limited. Interobserver reproducibility was assessed using a second medical officer who was trained on radiographic evaluation according to the Lane atlas. On random evaluation of 25% of the radiographs were reported by the second medical officer who was blinded to the first reader’s interpretations.

Multivariable analysis was performed using different regression models considering the nature of the outcome/dependent variables. Multivariable generalised linear model with linear response was used when the severity of disability and intensity of pain were used as the continuous outcome variables. X-ray features of LDD (disc space narrowing, anterior osteophytes and overall LDD) and presence of lumbar spondylolisthesis were defined as main independent variables/predictor variables and were treated as categorical variables. Separate linear regression models were created for each feature. In each multivariable generalised linear model with linear response, the magnitude of the association was presented as β coefficients with 95% confidence intervals (CI). Multivariable ordinal logistic regression was used when the severity of x-ray features of LDD (disc space narrowing, anterior osteophytes and overall LDD) were used as the ordinal outcome variables (0, 1 and 2). Multivariable logistic regression analysis was used when the presence of lumbar spondylolisthesis (yes/no) was used as a binary outcome variable. Magnitude of the associations was presented as adjusted odds ratios (aOR) with 95% CI in logistic regression models. Age, gender, BMI and presence of pain radiating into legs were defined as confounder variables in all regression models. Age and BMI were treated as continuous variables and gender (male/female) and presence of pain radiating into legs (yes/no) were treated as categorical variables.

Assumptions of ANOVA, independent samples t-test and regression models were verified. P value < 0.05 was used as the level of significance. Statistical analysis was carried out using SPSS version 17.

Table 1 summarises the characteristics of the study participants. Among 689 patients with chronic mechanical low back pain, 439 patients were recruited according to eligibility criteria. Thirteen patients had missing data for the variable BMI. Mean age ± SD was 48.99 ± 11.21 and 323 (73.58%) were females. BMI ± SD was 26.39 ± 4.65 and 87 (19.82%) were obese. Mean severity of disability was 30.95 ± 13.67 and mean intensity of pain was 45.50 ± 20.37. In addition, 110 (25.10%) patients had pain radiating into legs. With regard to interobserver reproducibility, intra-class correlation coefficient (ICC) of two readers for disc space narrowing was 0.88 (0.82-0.91) and ICC for anterior osteophytes was 0.81 (0.75 – 0.85). Among patients, 176 (40.09%) had disc space narrowing and 201 (45.78%) had anterior osteophytes with 69 (15.72%) and 26 (5.92%) having grade 2 disc space narrowing and grade 2 anterior osteophytes, respectively. LDD was present in 275 (62.64%) and 85 (19.36%) had grade 2 LDD. Lumbar spondylolisthesis was present in 51 (11.62%) patients.

There were no significant differences in severity of disability with the severity of disc space narrowing, anterior osteophytes and LDD according to ANOVA and generalised linear models with linear response (Table 2 and 3). Patients with the presence of lumbar spondylolisthesis had significantly severe disability in contrast to the patients without lumbar spondylolisthesis in both univariable and multivariable analysis (Table 2 and 3). Female gender and presence of pain radiating into legs were significantly associated with the severity of disability in all the multivariable generalised linear models (Table 3).

Disc space narrowing and LDD were not associated with intensity of pain in either univariable or multivariable regression analyses (Table 2 and 3). However patients with grade 1 anterior osteophytes had significantly higher intensity of pain compared to the patients with grade 0 anterior osteophytes. The presence of lumbar spondylolisthesis was not associated with the intensity of pain. Female gender and pain radiating into legs were associated with the intensity of pain in all multivariable generalised linear models (Table 3). In addition increasing age was associated with two linear regression models involving anterior osteophytes and LDD (Table 3).

The presence of grade 2 disc space narrowing was reported from 20 – 29 years age group, but presence of grade 2 anterior osteophytes was reported from 40 – 49 years age group. Furthermore the presence of lumbar spondylolisthesis was reported from 30 – 39 years age group. Advancing age was strongly associated with the severity of disc space narrowing, anterior osteophytes, LDD (Table 4) and presence of lumbar spondylolisthesis (aOR 1.15; 95% CI: 1.1 – 1.21) after adjusting for gender, BMI and pain radiating into legs. Male gender was associated with the severity of anterior osteophytes, but was not associated with disc space narrowing, overall LDD and lumbar spondylolisthesis. Furthermore, BMI was significantly associated with grades of anterior osteophytes and LDD.