Background

The cumulative or repetitive injury model was once a dominant paradigm of spine degeneration [1]. Thus, heavy occupational physical loading activities have long been suspected of increasing spine degeneration. However, inconsistencies between study findings, with some supporting this association [2, 3] and other not [4, 5], have led to controversy and uncertainty about the relationship between physical loading and lumbar spine degeneration. Furthermore, recent studies suggest that the structures of the spinal column, including the intervertebral discs, adapt and may even benefit from greater routine physical loading [1].

Controversy still exists between the relationship of occupational load and low back pain [6]. However, given the subjective nature of pain evaluation and the high prevalence of back pain in general, studies depicting the association between pain and occupation load always have large room for bias. The use of objective measures of spine degeneration to evaluate the impact of occupational load on the spine can provide a solution to better understanding this relationship. The evaluation of spine degeneration on imaging is both a reliable and objective measure to evaluate the effects of repetitive load on the spine, which in turn may mediate the occurrence of back pain in this population. Although spine degeneration on imaging is not synonym of back pain, spine degeneration on imaging have been found to be associated with an increased risk for low back pain [7] and increased risk of recurrent episodes [8].

Given the inconsistencies in the literature about the association of occupational load and spine degeneration, the objective of this study was to systematically review the literature on the association of occupational loading and spine degeneration observed on imaging. Occupational loading was described as loading conditions occurring during occupational activities, such as lifting and manual handling or comparisons between specific occupations.

Methods

A protocol for the study was developed a priori following the PRISMA guidelines and Cochrane Handbook.

Data sources and searches

A computerised electronic search was performed to identify relevant articles. The search was conducted on MEDLINE (1946 to May 2019), CINAHL (1982 to May 2019) and EMBASE (1988 to May 2019). Key words included in our search were related to 3 domains: imaging (i.e. x-ray, radiograph), imaging findings (i.e. disc degeneration, disc height) and load (i.e. manual handling, occupational load). Subject subheadings and word truncations specific for each database were used. There was no language restriction. See Additional file 1 for search strategy section.

Two reviewers screened search results (titles and abstracts) for potentially eligible studies. A third independent reviewer resolved any disagreement for inclusion of trials. Authors were contacted if more information about the trial was needed to allow inclusion of the study.

We also performed a search on the reference lists of the included studies and a search on ISI Web of Sicence (May 2019) for papers that cited the included studies.

Study selection

Cross-sectional, case-control and cohort studies evaluating occupational loading as the exposure were eligible for inclusion. All studies that evaluated professional athletes and whole body vibration as a form of exposure were included on a separate review. In addition, the study had to evaluate the relationship of loading with lumbar spine structural findings evaluated on diagnostic imaging. Studies that used back pain as an outcome measure were not included. Studies that included patients with pre-existing conditions, such as disc herniation, were excluded from the review as they are more likely to have positive findings on imaging and may provide biased estimates for the relationship under investigation. Two reviewers screened the full text of potentially eligible studies and decided on inclusion. A third independent reviewer resolved any disagreement for inclusion of studies. The reviewers followed a research protocol developed prior to the beginning of the review, which included a checklist of inclusion criteria.

Data extraction and quality assessment

The methodological quality of the trials was assessed using the Newcastle Ottawa Quality Assessment scale [9, 10] for case-control and cohort studies. The maximum value of the scale is 9 (high quality) and the minimum value is 0 (lowest quality). The quality was assessed by independent raters and disagreements were resolved by a third rater. Methodological quality was not an inclusion criterion but was taken into consideration when making conclusions.

Two independent reviewers (LM and research assistants) extracted data from the included studies using a standardized data extraction form. Important characteristics of each study were extracted, such as type of loading, study design, type of imaging, patient population, affiliation of the authors, funding source, and study conclusions. We also extracted the type of outcomes used, and for continuous outcome measurements we extracted mean scores, standard deviations and sample size, and for dichotomous and ordinal outcomes, sample size and number of events per group.

Data synthesis and analysis

Results were pooled when trials were considered sufficiently homogenous with respect to participant characteristics, exposure and outcomes. I2 was calculated using RevMan 5 to assess statistical heterogeneity. A random effects model was used to pool all available outcomes. I2 was calculated to evaluate statistical heterogeneity of pooled outcomes [11]. When adequate data were presented from the original study, mean differences and standard deviations for continuous outcomes and odds ratios for dichotomous outcomes were calculated. When such information was not available, the information presented in each study was used for interpretation of the results.

The GRADE approach for grading the level of the evidence available was used to summarize the conclusion of this review [11]. Depending on the number and quality of the studies included in the review, the evidence was classified into high, moderate, low or very low quality evidence.

Results

Study selection

The electronic database search resulted in a total of 5363 articles after removing duplicates. Of these, 137 were selected as potentially eligible based on their title and abstract.

After full title screening a total of 16 studies were included in the review. An additional ISI web of science search showed 11 more potentially eligible studies, from which 1 was included in the review. Therefore, the final number of included studies was 17. (Flowchart_ Fig. 1).

Fig. 1
figure 1

Flow chart of occupational load systematic review inclusion

Full size image

125 Exclusion for graph only * 27 No outcome of interest [6, 10, 12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30, 31,32,33,34,35], 11 no non-exposed group [36,37,38,39,40,41,42,43,44,45,46], 23 not evaluating occupational load [1, 47,48,49,50,51,52,53, 54,55,56,57,58,59,60,61,62], 19 not a cohort study [63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81], 11 included patients on already established conditions, [37, 82,83,84,85,86,87,88,89,90,91] 5 no imaging on all groups [92,93,94,95,96], 4 cervical spine [97,98,99,100], 1 in vitro, [101] 1 child study [34], 1 cant get full text [102], 3 secondary analysis or study with same population of an already included study, [103,104,105] 10 studies evaluated whole body vibration alone, [106,107,108,109,110,111,112,114,115] 9 studies evaluated athletes [116,17,18,19,20,21,22,23,124].

Study characteristics

There were 16 original studies evaluating occupational load [2,3,4,5, 125,126,127,128,129,130,131,132,133,134,135,136] including 1 follow-up study [137]. Ten studies evaluated specific job requirements such as occupational lifting or job load summary score [2, 4, 127,128,129,130, 132, 133, 135,136,137] and six compared different types of occupation with different occupational requirements [3, 5, 125, 126, 131, 134]. See Table 1 for study characteristics.

Table 1 Study characteristics
Full size table

Methodological quality

The methodological quality demonstrated an overall moderate level of quality, with a minimum of 3, a maximum score of 8, and a median and interquartile range of 5 and 3. The items of the methodological quality scale that were not present in most studies were control of potentially confounding factors and reporting of the response rate of each group.

Three studies included the same population from the Finnish Twin Spine study, but answered different questions related to different outcomes. These studies represent the strongest form of evidence given that controls are identical twins, minimizing possible confounding and familial aggregation [127, 128].

Outcomes

Ten studies evaluated the impact of loading on disc degeneration, generally assessed through disc signal intensity, representing disc desiccation [2,3,4,5, 126,127,128,129, 133, 134, 137]. The primary method of evaluating disc degeneration is through observing disc signal intensity on imaging. Disc degeneration is often associated with a whiter less translucent appearance of the disc [126]. Of these 10 studies only two did not find significant differences between groups [3, 129] and one study found more degeneration in those with less load [127]. In one study we were not able to assess whether statistical differences existed [134]. Seven studies identified some significant difference between loading groups with more load being associated with more degeneration, although we were not able to pool the results given the differences in types of loads and outcomes measured [2, 4, 5, 126,127,128, 133]. For all comparisons odds ratios when calculated varied between 1.89 to 3.7. A summary of the findings is presented in Table 2. One additional study looked at an overall measure of degeneration that included a combination of factors, and found that occupational loading was associated with the overall degeneration measurements [130].

Table 2 Exposure and results of each study included in the review that evaluated occupational load
Full size table

We were able to pool the results of 4 studies evaluating the association of disc degeneration with different types of occupational load for different spine levels [2,3,4, 133]. The results demonstrated that for all levels evaluated, including L1-S1, there was a statistically significant difference between loading groups with more degeneration associated with greater loading. Forest plost are presented in Fig. 2.

Fig. 2
figure 2

Disc degeneration (signal intensity) forest plots for each spinal level, L1-S1

Full size image

Seven studies evaluated disc height [2, 125, 127, 128, 131, 133, 137]. Disc height can be measured on imaging using quantitative or qualitative measures and is a surrogate measure of disc degeneration. Of the seven studies evaluating disc height only one did not find a significant difference in disc height between groups [2]. The other six studies identified some type of influence of occupation load with disc height, with greater load being associated with narrower discs. Four studies identified an overall relationship of loading with disc height, without focusing on specific levels [125, 127, 128, 137] and the other studies found different levels to be significantly different [131, 133].

Seven studies evaluated a difference in the prevalence of disc bulge or herniation [3, 126, 128, 130, 133, 136, 137]. Disc bulges or herniations were primarily evaluated through visual observation of images. Of the seven studies, five identified a significant difference between loading groups [3, 128, 133, 136, 137]. Three studies evaluated the prevalence of all lumbar levels together [128, 136, 137] while two studies found difference for different levels, [3, 133] which varied between the studies. When calculated, odd ratios varied between 2.0 to 3.1.

Three studies evaluated Modic changes [126, 129, 135]. Modic changes represent lesions of the vertebral endplate that is adjacent to the bone marrow. Modic changes are often assessed qualitatively [135]. In this review only one study [135] identified a relationship of modic changes with occupation load. One study evaluated the prevalence of Schmorl’s nodes [132]. These are small protrusion of the disc into the vertebral body. The one study included in this review did not identify a relationship of nodes with occupational loading groups. Two studies evaluated the presence of other endplate abnormalities, [125, 137] with only one study identifying a difference between loading groups [137]. Finally, three studies evaluated the presence of osteophytes [125, 132, 137] with two studies identifying greater prevalence of osteophytes in those with greater load [125, 137].

Discussion

The results of this study suggest that there is moderate grade evidence of an association between occupational loading and disc degeneration in terms of signal intensity. There is low quality grade evidence between loading and disc height, with inconsistent results between levels. There is low quality evidence for an association of disc bulging with occupational loading, again with inconsistent results among spinal levels. There is low quality evidence of an association between occupational loading and osteophytes, Modic changes, Schmorl’s nodes and other endplate abnormalities.

The results do suggest that occupations with greater physical loading are associated with modestly greater spine degeneration although differences in loading conditions and outcomes between studies make it is difficult to draw strong, specific conclusions. This is especially true given that positive results were inconsistently found at different spinal levels and for different outcomes. Thus, it remains difficult to draw conclusions about which type of loading may negatively affect which type of degenerative or structural findings. Additionally, different imaging methods were used (e.g. MRI, CTScans and x-rays) and different methods to assess spine degeneration make it difficult to draw conclusions.

Limitations of the review are primarily related to the heterogeneity of the studies included. There was a wide range of types of occupational loading studied and a wide range of outcomes evaluated. Thus, although odds ratios were presented in the original manuscripts for most of the studies, it was not possible to pool the great majority of results and the findings of the review, therefore, were presented qualitatively. Finally, the poor methodological quality of some of the studies, with only a small portion assessing degeneration longitudinally, limits interpretation regarding the progression of spine degeneration.

Future research should focus on more longitudinal studies, where the development of spinal degeneration can be followed over time, with an adequate follow-up period to allow for structural changes to occur. Monozygotic twin studies should be considered, given the strength of twin study designs in minimizing possible confounding. Furthermore, individual loading exposures should be taken in consideration, especially as the activities and loading involved in any one profession can vary significantly. More specifically, the type and magnitude of loading should be depicted and evaluated in greater detail. Finally, with the advance of imaging techniques and measurement procedures, a wide variety of measures of spinal degeneration and pathology has resulted. Guidelines for measurement and better standardization of spine imaging phenotypes are needed to allow study comparisons and pooling of data to facilitate interpretation of the collective body of related research.

Conclusion

The results of this study suggest that there is moderate grade evidence of an association between occupational loading and disc degeneration in terms of signal intensity (disc degeneration). There is low or very low-quality grade evidence between loading and disc height, disc bulging, osteophytes, Modic changes, Schmorl’s nodes and other endplate abnormalities. While there seems to be a modest association between heavy occupational loading and spinal degenerative findings, the limitations of the results found in this review provide a weak foundation for practical applications and related health policies.