Back pain is a near-universal human experience at some time during life, and neck pain is also common. The overwhelming majority of low back and cervical pain is considered to be due to unspecified mechanical factors or disc degeneration, which is a common with ageing and, hence, in people of working age. Back pain and disc disease appear to have significant heritability, based upon twin studies, but environmental factors also contribute – including physical occupational activities in some studies – although the strength of this association remains uncertain. This article examines the contribution of genetic and environmental factors to back pain and disc disease, with a specific focus on occupational exposures.
Back pain is highly prevalent in the Western world with a lifetime prevalence reaching up to 80%, and accounts for considerable work absenteeism. Low back pain (LBP) has been studied more extensively than neck pain, and a number of studies have shown a relationship between LBP and lumbar disc degeneration (LDD), although the strength of this association remains unclear . Following a general description of the epidemiology of cervical and lumbar pain and degenerative disc disease, this article focusses on the role of occupational factors in causation of cervical and lumbar pathology and factors that might mitigate risk. Areas of uncertainty and research interest are emphasised, where relevant. Attention is given to risk factors at the workplace and prevention of work-associated cases.
Some degree of LDD is almost universal in adults with ageing , and has been repeatedly proposed as one of the main causes of LBP . Although an association with LDD was first clearly demonstrated in twin studies , and more recently confirmed in a systematic review , specific risk factors for LBP remain unclear and universal consensus regarding the strength of the LDD–LBP relationship has been variable . In addition to occupation, other environmental factors, such as obesity and smoking, have been reported to be associated with prevalent LBP, although the quantitative effect of the majority of these has been found to be small . On the other hand, several twin studies have consistently suggested the presence of a significant genetic component underlying variation in both LBP and LDD .
Clinical features
Classically, the syndrome of LBP related to LDD comprises somatic referred pain with or without radiculopathic features, together with abnormalities on imaging.
However, because neck and back pain are intermittent symptoms, a number of studies have reported only relatively modest relationships between back pain and abnormal radiology whether imaged by plain X-rays, computed tomography (CT) scanning or magnetic resonance imaging (MRI). A recent systematic review suggests that the odds ratio (OR) for an association with LBP is strongest for disc protrusion (OR 3.6), followed by disc degeneration (OR 2,5), annular tears (OR 2.5) and nerve root compression/displacement (OR 2.3).
Case definitions and diagnosis
The lack of standardised clinical criteria and radiological definitions has hampered the undertaking of well-executed epidemiological studies. When LBP is defined by severity of symptoms alone rather than disability as well, there appears to be considerable heterogeneity in the underlying aetiology . More sensitive imaging modalities, especially MRI, have become gradually more widely available and facilitated larger-scale epidemiological studies of LDD. A number of studies have provided estimates of the association between disc degeneration and LBP .
Case definitions and diagnosis
The lack of standardised clinical criteria and radiological definitions has hampered the undertaking of well-executed epidemiological studies. When LBP is defined by severity of symptoms alone rather than disability as well, there appears to be considerable heterogeneity in the underlying aetiology . More sensitive imaging modalities, especially MRI, have become gradually more widely available and facilitated larger-scale epidemiological studies of LDD. A number of studies have provided estimates of the association between disc degeneration and LBP .
Epidemiology
The traditional view of intervertebral disc degeneration (IDD) has been that it was a process related to normal ageing as well as changes related to physical loading over the lifetime. Fundamental to any description of the epidemiology of IDD, however, is the definition of a case and how to measure these. There is no standard definition of back pain or disc degeneration and, for this reason, comparison between studies is difficult. Operationally, disc degeneration is defined largely by the method of evaluation. For example, radiography or CT scan can most usefully assess disc height and osteophytes, whereas MRI can better assess disc signal and structural change, such as prolapse or herniation. For large population studies, the preferred method is MRI, and most systems of evaluation although qualitative, include assessment of disc height, signal intensity, bulging or prolapse and osteophytes. The prevalence rates of these measures differ considerably between studies. For example, the prevalence of lumbar disc bulge ranges from 10% to more than 80% in so-called asymptomatic subjects, and the prevalence of annular tears from 6% to 56% . The prevalence of protrusion ranges from 4% to 76%, and the prevalence of disc degeneration, usually defined by the presence of disc height loss and/or reduced T2 signal intensity, is estimated at 54% .
There also appears to be variation in the prevalence of MRI changes at the five different lumbar disc levels with, in general, the lower levels having the highest prevalence of disc changes ( Fig. 1 ), with the exception of Schmorl’s nodes, which are more common in the upper lumbar and thoracic region . For these reasons, there is debate regarding the use of summary scores to describe LDD, which have been advocated by some but not others .
As noted above, there is no agreement regarding the most appropriate definition of LBP and neck pain for use in population studies. Definitions that do not take into account pain intensity or disability may overestimate pain that is of little or no public health importance. McGregor et al. focussed on pain with a duration of greater than 1 month, accompanied by distal radiation and associated with disability. This definition of pain related to the lifetime prevalence of pain (that is report of pain “ever”) rather than to the 1-year period prevalence, which is more commonly used in case-control studies. For LBP using this more conservative definition, the lifetime cumulative prevalence of duration longer than a month associated with severe disability was 14% . The prevalence of neck pain, using an equivalent definition, was 8%.
By contrast, a systemic review that used more liberal definitions has estimated the 1-month prevalence of LBP to be 30% and the 1-year prevalence to be 50%, and the combined prevalence from three population studies was 67% .
The progression of LDD over time has been investigated by Videman et al. , who investigated 140 male monozygotic (MZ) twins, and studied a variety of MRI traits over 5 years. They observed that there were slow changes in LDD over time with the different subtraits progressing at different rates. Borenstein et al. performed MRI scans on 50 subjects, who had no history of back pain, as baseline (mean age 43.6 years) and repeated the MRI 7 years later in 31 subjects (mean age at follow-up 52 years). Over 7 years, 21 subjects developed back or leg pain, including 12 with a normal scan, five of six subjects with a herniated disc and three of four subjects with spinal stenosis. With regard to the ability of MRI to predict LBP, a positive trend was noted in each disease category, but none was significant; however, with this small sample, the study was underpowered to a large extent. Hassett et al. reported progression rates of 3.9% for anterior osteophytes and 3.2% for disc- space narrowing using plain radiography in 914 female volunteers . The UK Twin Spine Study has performed a 10-year follow-up on 460 twins, and has shown that while all MRI subtraits deteriorated significantly over time, there appears to be little influence on the episodes of severe and disabling back pain reported at the two time points.
Genetic studies
Twin studies have provided considerable information about the role of genetic influences on LDD. Battie et al. studied 115 identical male twins selected for discordance in suspected environmental risk factors, such as smoking. In multivariate analyses, job code explained 7% of the variability in the upper lumbar (T12–L4) degenerative summary score, the addition of age explained a further 9% and when twinship was added, the amount of variability explained rose to 77%. In the lower lumbar spine, heavy leisure-time physical loading explained only 2% of the variability, age 9%, and with twinship, 43% of variability was explained. Although these studies in identical twins suggest a genetic effect, a strong effect of the family environment cannot be discounted. Unfortunately, non-identical twins were not included in the study, making interpretation of the heritability impossible.
To determine heritability of LDD, Sambrook et al. used a cohort of mainly female UK and Australian twins unselected for back pain, and reported that overall heritability was 74% for LDD and 73% for cervical disc degeneration. Examination of individual phenotypic features revealed that disc height and bulge were highly heritable at both sites (59–79%), with osteophytes heritable in the lumbar spine (59%). Osteophytes were not shown to be heritable in the cervical spine, but it must remembered that MRI is not the optimal imaging modality for assessing bony changes, and cervical spine MRI images are smaller than the lumbar spine images, making accurate assessment of this subtrait a challenge. It must also be recognised that twin studies can sometimes produce an exaggerated estimate of heritability due to certain biases, most commonly selection bias, whereby subjects with symptoms self-select for a study, or due to an imbalance in environmental factors common to a pair, identical twins have greater similarity in environmental factors that are risk factors for a disease. Sambrook et al. showed there was greater concordance in identical twins for certain factors, such as occupational manual work and smoking; however, adjustment for these did not diminish the estimates of heritability, suggesting their effects were small. On the other hand, the number of males in the study was too small to assess definitely any differences between males and females; hence, some caution may be necessary in extrapolating these results, to assess occupational effects in men.
With regard to back pain, two earlier questionnaire-based surveys conducted in the Nordic twin registries, using non-standard case definitions, had produced conflicting results. A relatively small genetic influence on LBP was observed by Heikkila et al. in a Finnish survey for ‘sciatica’ diagnosed by a physician (where heritability was estimated at 21%) . A larger genetic contribution (heritability 50%) was observed in a survey conducted in Finland for a single question relating to back pain resulting in absence from work . Neither study included objective measures of disc degeneration or covariates.
Using the same, large, unselected, twin population studied as Sambrook et al. , McGregor et al. reported a significant genetic effect on LBP and neck pain with estimates of heritability ranging from 52% to 57% for LBP and from 35% to 48% for neck pain. The sources of variation were complex, with major contributions from structural MRI changes for LBP (OR 3.63 for upper quartile of MRI grade), body mass index (OR 2.34 for upper quartile) and smoking (OR 1.60 for “ever smoking”). In this study, MRI change was the strongest single predictor of LBP, whereas for neck pain, psychological distress, smoking and age were more important .
Williams et al. have studied Schmorl’s nodes in twins and found them in 30% of subjects at any vertebral level, with multiple Schmorl’s nodes in 14% (with equal to or more than two nodes). In 9288 vertebral end plates, 374 Schmorl’s nodes were found, 153 (41%) in the lumbar spine and 221 (59%) in the thoracic spine. The heritability of Schmorl’s nodes was over 70%. There was a positive association between Schmorl’s nodes and LDD. Schmorl’s nodes were more frequent in subjects with back pain, but this association was explained by the association of Schmorl’s nodes with LDD, and Schmorl’s nodes were not themselves an independent risk factor for back pain.
A number of subsequent studies have studied candidate genes for LDD, and have identified a number of variants, including Vitamin D receptor (VDR); collagen, type IX, alpha 2 (COL9A2); collagen, type IX, alpha 3 (COL9A3); collagen, type I, alpha 1 (COL1A1); matrilin3 (MATN3); and matrix metalloproteinase (MMP), which have been replicated in some studies, and some inflammatory genes, such as IL-1 and THSD2 . However, little is known of the genetic or environmental factors governing the rate of progression of LDD.
Williams et al. have recently reported linkage studies in 348 women from the Twins UK register, who had undergone spine MRI scanning 10 years earlier. Significant linkage peaks (defined as maximum log of odds (LOD) > 3) were identified for LDD at three chromosomal regions. These included chromosome 1 (position 285 centimorgan (cM)), chromosome 5 (position 175 cM) and chromosome 19 (position 80 cM), close to a peak previously obtained for hand osteoarthritis. The peak on chromosome 19 had an LOD score of 4.06, and the empirical p = 6.7 × 10 −4 confirmed reliability of the linkage signal.
Solovieva et al. studied 135 middle-aged men, and reported that the Trp3 allele of the COL9A3 gene plus obesity acted synergistically to increase the risk of a dark nucleus pulposus and posterior disc bulge, and decreased disc height on MRI. They estimated that 45–71% of disc degeneration in persistently overweight individuals with the Trp 3 allele could be attributed to this interaction. These investigators further examined the association between collagen and IL-1β gene polymorphisms and LDD in these same 135 middle-aged occupationally active men. The Trp 3 allele of the COL9A3 gene increased the risk of signal loss on MRI in the absence of the IL-1 βT allele (OR 7) and degenerative changes (OR 8); however, there was no effect of the Trp3 allele in the presence of the IL-1 βT allele. Carriers of the Col11α2 minor allele also had a modest increased risk of disc bulges (OR 2.1) as compared with non-carriers.
In a recent review, Zhang et al. concluded that LDD was not only regulated by multiple genes but also by environmental factors, with gene–gene and gene-environment interactions likely to be present. They concluded that although genome-wide screening had identified a number of genes, each gene appeared to make only a small overall contribution to explaining LDD, and that long-term studies with larger sample sizes were needed.
Occupational associations
In the UK, approximately 20 million working days are lost each year because of LBP, and the latter accounts for 40% of the time lost due to industrial injury .
Although the best evidence for assessing the influence of risk factors on disease comes from population-based, prospective, cohort studies, with risk exposures determined at baseline, in practice, a case-control design is more frequently used, as this approach is more feasible. In such a study design, accurate retrospective assessment of occupational exposure is difficult, and usually relies on self-reporting of such exposures rather than objective measures. The other problem in assessing the relationship between occupational exposure and LDD is that there is likely to be a prolonged asymptomatic lead time and, indeed, back pain is usually an intermittent phenomenon – unlike the radiological changes – which are gradually progressive. In these studies, occupational exposure has generally been assessed by categories of activity, for example, heavy lifting, frequent bending or broad occupational categories, such as clerical versus heavy manual labour or truck driving. End points examined have ranged from symptomatic back pain to radiological change or surgical end points for disc prolapse.
One of the most comprehensive studies of the relationship between LBP and occupation was conducted by Lotters et al. The authors performed a systematic review of the literature and calculated the pooled prevalence of LBP in an unexposed population and the pooled OR for each work-related risk factor in a meta-analysis, using a random effects model. An unbiased risk estimate for each risk factor was obtained by correcting the pooled OR for confounding by other risk factors. The pooled prevalence for LBP among unexposed subjects was 22%, 30% and 34% for the <35-year, 35–45-year and >45-year age categories, respectively. The pooled OR was 1.51 (95% confidence interval (CI) 1.31–1.74) for manual materials’ handling, 1.68 (95% CI 1.41–2.01) for frequent bending or twisting, 1.39 (95% CI 1.24–1.55) for whole-body vibration and 1.30 (1.17–1.45) for job dissatisfaction. For high exposure to manual materials’ handling, frequent bending or twisting and whole-body vibration, the pooled OR was 1.92, 1.93 and 1.63, respectively. They proposed an algorithm for calculating the level of work-relatedness of LBP using the aetiological fraction ( Fig. 2 ). To determine the likelihood of work-relatedness for LBP dichotomously, they proposed a cut-off point of 50%, meaning that, if 50% or more of the calculated probability was due to occupational exposure, the LBP could be regarded as work related, noting that an aetiological fraction of 50% is often used in decision making, for example, in compensating lung cancer patients occupationally exposed to asbestos.
Related posts:
Hand pain other than carpal tunnel syndrome (CTS): The role of occupational factors
Shoulder pain at the workplace
Hip and knee pain: Role of occupational factors
Factors that affect the occurrence and chronicity of occupation-related musculoskeletal disorders
Epidemiological aspects of studying work-related musculoskeletal disorders
Management of occupation-related musculoskeletal disorders
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
