Advertisement

The Effect of Smoking on the Risk of Sciatica: A Meta-analysis

Published:September 21, 2015DOI:https://doi.org/10.1016/j.amjmed.2015.07.041

      Abstract

      Background

      The role of smoking in sciatica is unknown. This study aimed to estimate the effect of smoking on lumbar radicular pain and clinically verified sciatica.

      Methods

      Comprehensive literature searches were conducted in PubMed, Embase, Web of Science, Scopus, Google Scholar, and ResearchGate databases from 1964 through March 2015. We used a random-effects meta-analysis, assessed heterogeneity and publication bias, and performed sensitivity analyses with regard to study design, methodological quality of included studies, and publication bias.

      Results

      Twenty-eight (7 cross-sectional [n = 20,111 participants], 8 case control [n = 10,815], and 13 cohort [n = 443,199]) studies qualified for a meta-analysis. Current smokers had an increased risk of lumbar radicular pain or clinically verified sciatica (pooled adjusted odds ratio [OR] 1.46; 95% confidence interval [CI], 1.30-1.64, n = 459,023). Former smokers had only slightly elevated risk compared with never smokers (pooled adjusted OR 1.15; 95% CI, 1.02-1.30, n = 387,196). For current smoking the pooled adjusted OR was 1.64 (95% CI, 1.24-2.16, n = 10,853) for lumbar radicular pain, 1.35 (95% CI, 1.09-1.68, n = 110,374) for clinically verified sciatica, and 1.45 (95% CI, 1.16-1.80, n = 337,796) for hospitalization or surgery due to a herniated lumbar disc or sciatica. The corresponding estimates for past smoking were 1.57 (95% CI, 0.98-2.52), 1.09 (95% CI, 1.00-1.19), and 1.10 (95% CI, 0.96-1.26). The associations did not differ between men and women, and they were independent of study design. Moreover, there was no evidence of publication bias, and the observed associations were not due to selection or detection bias, or confounding factors.

      Conclusions

      Smoking is a modest risk factor for lumbar radicular pain and clinically verified sciatica. Smoking cessation appears to reduce, but not entirely eliminate, the excess risk.

      Keywords

      Clinical Significance
      • Smoking is a modest risk factor for lumbar radicular pain and clinically verified sciatica.
      • Smoking cessation appears to reduce, but not entirely eliminate, the excess risk.
      • The association between smoking and lumbar radicular pain or clinically verified sciatica does not differ between men and women.
      Sciatica is lumbar pain that radiates to the leg along the sciatic nerve and is accompanied by clinical findings suggestive of nerve root irritation.
      • Koes B.W.
      • van Tulder M.W.
      • Peul W.C.
      Diagnosis and treatment of sciatica.
      Sciatica is the most common symptom of a herniated lumbar disc.
      • Koes B.W.
      • van Tulder M.W.
      • Peul W.C.
      Diagnosis and treatment of sciatica.
      The prevalence of clinically verified sciatica ranges between 2.2% and 5.1% in the general population.
      • Younes M.
      • Bejia I.
      • Aguir Z.
      • et al.
      Prevalence and risk factors of disk-related sciatica in an urban population in Tunisia.
      • Leino-Arjas P.
      • Kauppila L.
      • Kaila-Kangas L.
      • Shiri R.
      • Heistaro S.
      • Heliövaara M.
      Serum lipids in relation to sciatica among Finns.
      • Heliövaara M.
      • Mäkelä M.
      • Knekt P.
      • Impivaara O.
      • Aromaa A.
      Determinants of sciatica and low-back pain.
      It ranges between 2.7% and 5.7% in men, and between 1.4% and 4.5% in women.
      • Younes M.
      • Bejia I.
      • Aguir Z.
      • et al.
      Prevalence and risk factors of disk-related sciatica in an urban population in Tunisia.
      • Leino-Arjas P.
      • Kauppila L.
      • Kaila-Kangas L.
      • Shiri R.
      • Heistaro S.
      • Heliövaara M.
      Serum lipids in relation to sciatica among Finns.
      • Heliövaara M.
      • Mäkelä M.
      • Knekt P.
      • Impivaara O.
      • Aromaa A.
      Determinants of sciatica and low-back pain.
      The role of personal risk factors in sciatica is partly known.
      • Koes B.W.
      • van Tulder M.W.
      • Peul W.C.
      Diagnosis and treatment of sciatica.
      Of lifestyle risk factors, obesity is a modest risk factor for lumbar radicular pain and clinically verified sciatica.
      • Shiri R.
      • Lallukka T.
      • Karppinen J.
      • Viikari-Juntura E.
      Obesity as a risk factor for sciatica: a meta-analysis.
      The effects of cigarette smoking and leisure-time physical inactivity on sciatica are not yet clear.
      • Shiri R.
      • Karppinen J.
      • Leino-Arjas P.
      • et al.
      Cardiovascular and lifestyle risk factors in lumbar radicular pain or clinically defined sciatica: a systematic review.
      We previously conducted a systematic review on the association between smoking and lumbar radicular pain or clinically verified sciatica.
      • Shiri R.
      • Karppinen J.
      • Leino-Arjas P.
      • et al.
      Cardiovascular and lifestyle risk factors in lumbar radicular pain or clinically defined sciatica: a systematic review.
      Based on a qualitative synthesis, 5 of 11 studies found an association between smoking and lumbar radicular pain or sciatica. Long-term smoking was associated with an increased risk of lumbar radicular pain in some occupational populations. We also conducted a meta-analysis on the effect of smoking on nonspecific low back pain.
      • Shiri R.
      • Karppinen J.
      • Leino-Arjas P.
      • Solovieva S.
      • Viikari-Juntura E.
      The association between smoking and low back pain: a meta-analysis.
      Current smokers as well as past smokers had an increased risk of nonspecific low back pain. Depending on the severity of nonspecific low back pain, the excess risk ranged between 10% and 30% for past smoking and between 20% and 60% for current smoking.
      • Shiri R.
      • Karppinen J.
      • Leino-Arjas P.
      • Solovieva S.
      • Viikari-Juntura E.
      The association between smoking and low back pain: a meta-analysis.
      To date, it is unknown whether smoking is a risk factor for lumbar radicular pain only, a subjectively assessed outcome, or it is also a risk factor for clinically verified sciatica, a more specific and objectively assessed outcome. The aim of this review was to carry out a meta-analysis to estimate the magnitude of the effect of smoking on lumbar radicular pain that was not verified by clinical examination or imaging findings, and lumbar radicular pain that was verified as being due to a herniated lumbar disc or clinically verified sciatica. To include studies published after our qualitative review, a period of 9 years, we conducted a comprehensive literature search and reassessed the previous studies regarding their eligibility for a meta-analysis.

      Methods

       Search Strategy

      The protocol and reporting of the results of this systematic review and meta-analysis were based on the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) statement.
      • Moher D.
      • Liberati A.
      • Tetzlaff J.
      • Altman D.G.
      • Group P.
      Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement.
      The first reviewer conducted comprehensive literature searches in PubMed, Embase, Scopus, Web of Science, Google Scholar and ResearchGate using predefined keywords (Supplementary Table 1, Appendix, available online). Both Medical Subject Headings (MeSH) terms and text words were used in PubMed, and both Emtree (Embase subject headings) terms and text words were used in Embase. There was no language restriction. The first reviewer hand-searched the reference lists of included articles for additional studies, and looked at the full text of studies on the associations of overweight/obesity and physical activity/inactivity with lumbar radicular pain or sciatica for additional studies on smoking.
      • Shiri R.
      • Karppinen J.
      • Leino-Arjas P.
      • et al.
      Cardiovascular and lifestyle risk factors in lumbar radicular pain or clinically defined sciatica: a systematic review.
      Moreover, the first reviewer looked at the full text of studies on the associations of overweight/obesity, smoking and physical activity/inactivity with low back pain to identify additional studies on lumbar radicular pain or sciatica.
      • Shiri R.
      • Karppinen J.
      • Leino-Arjas P.
      • Solovieva S.
      • Viikari-Juntura E.
      The association between smoking and low back pain: a meta-analysis.
      • Shiri R.
      • Karppinen J.
      • Leino-Arjas P.
      • Solovieva S.
      • Viikari-Juntura E.
      The association between obesity and low back pain: a meta-analysis.

       Inclusion and Exclusion Criteria

      The first reviewer assessed the titles, abstracts, and full texts of the studies and investigated whether the studies examined an association between smoking and lumbar radicular pain or clinically verified sciatica. We included cross-sectional studies, both population-based and hospital-based case control studies and cohort studies in the systematic review. To be eligible for a meta-analysis, the studies had to report quantitative data on the association between smoking and lumbar radicular pain or sciatica. We excluded studies that were conducted among self-selected volunteers. Volunteers are generally healthier than the general population and they are more likely to be nonsmokers. Including studies conducted among self-selected volunteers can result in an overestimation of excess risk due to smoking. Moreover, we excluded studies that recruited patients with lumbar radicular pain or sciatica with no control group, and studies that did not report quantitative results to estimate an effect size.

       Quality Assessment

      We independently rated the quality of included studies using criteria adapted from the Effective Public Health Practice Project tool for observational studies.
      • Armijo-Olivo S.
      • Stiles C.R.
      • Hagen N.A.
      • Biondo P.D.
      • Cummings G.G.
      Assessment of study quality for systematic reviews: a comparison of the Cochrane Collaboration Risk of Bias Tool and the Effective Public Health Practice Project Quality Assessment Tool: methodological research.
      We assessed 5 sources of bias: selection bias, performance bias, detection bias, confounding, and attrition bias (Supplementary Table 2, Appendix, available online). Discrepancies in the quality assessments between 2 reviewers were resolved through discussion.

       Meta-analysis

      We estimated a prevalence ratio for cross-sectional studies, an odds ratio for case control studies, and a risk ratio for cohort studies. Woolf
      • Woolf B.
      On estimating the relation between blood group disease.
      confidence intervals were calculated for the estimated odds ratios. A fixed-effect meta-analysis was used to combine the subgroups of a single study and a random-effects meta-analysis to combine the estimates of different studies.

      Higgins J, Green S. Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0. The Cochrane Collaboration; 2009 [updated March 2011]. Available at: www.cochrane-handbook.org. Accessed July 2015.

      One study
      • Chiang S.
      • Lin C.
      • Tsai S.
      • et al.
      Cigarette smoking dose as a predictor of need for surgical intervention in patients with lumbar disk herniation.
      reported the effect sizes for 2 outcomes, hospitalization for a herniated lumbar disc and discectomy for a herniated lumbar disc for the same study population. For this study we combined the 2 estimates into a single estimate using the formula
      • Borenstein M.
      • Hedges L.
      • Higgins J.
      • Rothstein H.
      Introduction to Meta-analysis.
      suggested for combining multiple outcomes within a single study. Cross-sectional, case control, and cohort studies reported similar results (Table), thus, all designs were combined in a single analysis.
      TableA Sensitivity Analysis of 26 Studies That Reported Estimates for Smoking Adjusted at Least for Age and Sex According to Study Design, Methodological Quality of Included Studies, and Adjustment for Publication Bias
      CharacteristicPast SmokingCurrent SmokingEver Smoking
      SampleOR95% CIP-ValueSampleOR95% CIP-ValueSampleOR95% CIP-Value
      Overall387,1961.151.02-1.30459,0231.461.30-1.64397,0231.351.24-1.48
      Adjustment for publication bias1.090.94-1.271.371.21-1.571.311.18-1.45
      Study design
       Cross-sectional11,3171.060.75-1.4913,9781.350.90-2.0212,1751.130.81-1.57
       Case control10181.100.75-1.6337601.811.04-3.1480731.421.21-1.67
       Cohort374,8611.191.02-1.39.62441,2851.431.26-1.61.76376,7751.371.21-1.54.41
      Selection bias
       Low378,5161.151.00-1.33390,2741.341.18-1.51380,0001.291.14-1.45
       Moderate/high86801.200.90-1.61.8668,7491.631.28-2.07.3817,0231.411.22-1.64.50
      Confounding
       Low376,0101.100.97-1.24389,7551.341.19-1.52376,8681.271.13-1.44
       Moderate11,1861.351.02-1.77.2269,2681.701.30-2.23.2020,1551.421.25-1.61.55
      Detection bias
       Low280,5971.090.97-1.23349,7631.411.18-1.69287,6521.351.19-1.53
       Moderate106,5991.271.01-1.60.82109,2601.551.29-1.86.61109,3711.381.18-1.62.61
      Attrition bias
       Low122,7811.091.02-1.17183,6051.521.31-1.76131,8431.331.21-1.46
       Moderate264,4151.860.90-3.82.43275,4181.361.02-1.80.52265,1801.681.07-2.63.52
      Meta-regression was used to test for the differences between subgroups.
      The presence of heterogeneity across the studies was assessed by the I2 statistics.
      • Higgins J.P.
      • Thompson S.G.
      Quantifying heterogeneity in a meta-analysis.
      The I2 statistic shows the total variation across studies, which is not due to chance. An I2 statistic <25% indicates small inconsistency, and more than 50% indicates large inconsistency.
      • Ioannidis J.P.
      • Patsopoulos N.A.
      • Evangelou E.
      Uncertainty in heterogeneity estimates in meta-analyses.
      The influence of each individual study on the summary estimate and heterogeneity was examined by repeating the meta-analysis with one study out at a time. Meta-regression
      • Morton S.C.
      • Adams J.L.
      • Suttorp M.J.
      • Shekelle P.G.
      Meta-regression Approaches: What, Why, When, and How? Technical Review 8.
      was used to explore, whether study-level covariates accounted for the observed heterogeneity and to test for differences in the effect size between 2 or more subgroups.
      We used a funnel plot for exploring publication bias, Egger's regression test for examining funnel plot asymmetry, and the trim-and-fill method for estimating the number of missing studies attributable to publication bias.
      • Rothstein H.
      • Sutton A.
      • Borenstein M.
      Publication Bias in Meta-analysis: Prevention, Assessment and Adjustments.
      • Duval S.
      • Tweedie R.
      Trim and fill: a simple funnel-plot-based method of testing and adjusting for publication bias in meta-analysis.
      Statistical significance for publication bias was based on a P value <.10.
      • Borenstein M.
      • Hedges L.
      • Higgins J.
      • Rothstein H.
      Introduction to Meta-analysis.
      We used Stata, version 13 (StataCorp LP, College Station, TX) for meta-analysis.

      Results

       Study Selection

      Electronic searches identified 7124 publications (Supplementary Figure 1, Appendix, available online). We looked at the full-text of 467 relevant publications on the association between smoking and low back pain or lumbar disc disorders. Of 62 eligible studies on lumbar radicular pain or clinically verified sciatica, we excluded 27 studies that were conducted among patients with sciatica with no control group, 3 on self-selected volunteers, 3 with insufficient quantitative data to estimate an effect size, and one study that matched cases and controls for smoking. Of the 3 excluded studies on volunteers, 2 also had another limitation. One study
      • Kanayama M.
      • Togawa D.
      • Takahashi C.
      • Terai T.
      • Hashimoto T.
      Cross-sectional magnetic resonance imaging study of lumbar disc degeneration in 200 healthy individuals.
      used a radiological lumbar disc herniation, either symptomatic or asymptomatic, as an outcome. Another study
      • Södervall J.
      • Karppinen J.
      • Puolitaival J.
      • et al.
      Heart rate variability in sciatica patients referred to spine surgery: a case control study.
      did not aim to examine the role of smoking in sciatica. Cases and controls differed in several important risk factors for low back disorders such as sex, body mass index, and physical activity. However, only an unadjusted estimate was reported.
      Finally, we included 7 cross-sectional (number of participants [n] = 20,111), 8 case control (n = 10,815), and 13 cohort (n = 443,199) studies on the association between smoking and lumbar radicular pain or clinically verified sciatica in the meta-analysis (Supplementary Table 3, Appendix, available online). There were 9 studies
      • Korovessis P.
      • Repantis T.
      • Zacharatos S.
      • Baikousis A.
      Low back pain and sciatica prevalence and intensity reported in a Mediterranean country: ordinal logistic regression analysis.
      • Matsudaira K.
      • Kawaguchi M.
      • Isomura T.
      • et al.
      Identification of risk factors for new-onset sciatica in Japanese workers: findings from the Japan epidemiological research of Occupation-related Back pain study.
      • Kääriä S.
      • Leino-Arjas P.
      • Rahkonen O.
      • Lahti J.
      • Lahelma E.
      • Laaksonen M.
      Risk factors of sciatic pain: a prospective study among middle-aged employees.
      • Leino-Arjas P.
      • Solovieva S.
      • Kirjonen J.
      • Reunanen A.
      • Riihimäki H.
      Cardiovascular risk factors and low-back pain in a long-term follow-up of industrial employees.
      • Miranda H.
      • Viikari-Juntura E.
      • Martikainen R.
      • Takala E.P.
      • Riihimäki H.
      Individual factors, occupational loading, and physical exercise as predictors of sciatic pain.
      • Manninen P.
      • Riihimäki H.
      • Heliövaara M.
      Incidence and risk factors of low-back pain in middle-aged farmers.
      • Riihimäki H.
      • Viikari-Juntura E.
      • Moneta G.
      • Kuha J.
      • Videman T.
      • Tola S.
      Incidence of sciatic pain among men in machine operating, dynamic physical work, and sedentary work. A three-year follow-up.
      • Riihimäki H.
      • Wickström G.
      • Hänninen K.
      • Luopajärvi T.
      Predictors of sciatic pain among concrete reinforcement workers and house painters—a five-year follow-up.
      • Karjalainen U.
      • Paananen M.
      • Okuloff A.
      • et al.
      Role of environmental factors and history of low back pain in sciatica symptoms among Finnish adolescents.
      on lumbar radicular pain that was not verified by clinical examination or imaging findings, and 19 studies on lumbar radicular pain that was verified as being due to a herniated lumbar disc or sciatica. Of the latter group, 8 studies were on clinically verified sciatica,
      • Younes M.
      • Bejia I.
      • Aguir Z.
      • et al.
      Prevalence and risk factors of disk-related sciatica in an urban population in Tunisia.
      • Leino-Arjas P.
      • Kauppila L.
      • Kaila-Kangas L.
      • Shiri R.
      • Heistaro S.
      • Heliövaara M.
      Serum lipids in relation to sciatica among Finns.
      • Heliövaara M.
      • Mäkelä M.
      • Knekt P.
      • Impivaara O.
      • Aromaa A.
      Determinants of sciatica and low-back pain.
      • Zhang Y.G.
      • Zhang F.
      • Sun Z.
      • et al.
      A controlled case study of the relationship between environmental risk factors and apoptotic gene polymorphism and lumbar disc herniation.
      • Kostova V.
      • Koleva M.
      Back disorders (low back pain, cervicobrachial and lumbosacral radicular syndromes) and some related risk factors.
      • Kelsey J.L.
      • Githens P.B.
      • O'Conner T.
      • et al.
      Acute prolapsed lumbar intervertebral disc. An epidemiologic study with special reference to driving automobiles and cigarette smoking.
      • Jhawar B.S.
      • Fuchs C.S.
      • Colditz G.A.
      • Stampfer M.J.
      Cardiovascular risk factors for physician-diagnosed lumbar disc herniation.
      • Violante F.S.
      • Fiori M.
      • Fiorentini C.
      • et al.
      Associations of psychosocial and individual factors with three different categories of back disorder among nursing staff.
      and 11 on hospitalization or surgery due to a herniated lumbar disc or sciatica.
      • Chiang S.
      • Lin C.
      • Tsai S.
      • et al.
      Cigarette smoking dose as a predictor of need for surgical intervention in patients with lumbar disk herniation.
      • Schumann B.
      • Bolm-Audorff U.
      • Bergmann A.
      • et al.
      Lifestyle factors and lumbar disc disease: results of a German multi-center case-control study (EPILIFT).
      • Zhang Y.G.
      • Sun Z.
      • Zhang Z.
      • Liu J.
      • Guo X.
      Risk factors for lumbar intervertebral disc herniation in Chinese population: a case-control study.
      • Heliövaara M.
      • Knekt P.
      • Aromaa A.
      Incidence and risk factors of herniated lumbar intervertebral disc or sciatica leading to hospitalization.
      • Sørensen I.G.
      • Jacobsen P.
      • Gyntelberg F.
      • Suadicani P.
      Occupational and other predictors of herniated lumbar disc disease-a 33-year follow-up in the Copenhagen male study.
      • Wahlström J.
      • Burstrom L.
      • Nilsson T.
      • Järvholm B.
      Risk factors for hospitalization due to lumbar disc disease.
      • Kaila-Kangas L.
      • Leino-Arjas P.
      • Riihimäki H.
      • Luukkonen R.
      • Kirjonen J.
      Smoking and overweight as predictors of hospitalization for back disorders.
      • Saftic R.
      • Grgic M.
      • Ebling B.
      • Splavski B.
      Case-control study of risk factors for lumbar intervertebral disc herniation in Croatian island populations.
      • An H.S.
      • Silveri C.P.
      • Simpson J.M.
      • et al.
      Comparison of smoking habits between patients with surgically confirmed herniated lumbar and cervical disc disease and controls.
      • Rivinoja A.E.
      • Paananen M.V.
      • Taimela S.P.
      • et al.
      Sports, smoking, and overweight during adolescence as predictors of sciatica in adulthood: a 28-year follow-up study of a birth cohort.
      • Mattila V.M.
      • Saarni L.
      • Parkkari J.
      • Koivusilta L.
      • Rimpelä A.
      Early risk factors for lumbar discectomy: an 11-year follow-up of 57,408 adolescents.

       Methodological Quality of Included Studies

      Twelve studies were rated as having low risk of selection bias, 13 as having moderate risk of selection bias, and 3 as having high risk of selection bias. Seventeen studies were rated as having low risk of detection bias and 11 as having moderate risk of detection bias. Sixteen studies differentiated between never, former, and current smokers, while 4 studies compared current smokers with never or past smokers, and 8 studies did not differentiate between current and former smokers.

       Smoking and Sciatica

      Current smokers had an increased risk of lumbar radicular pain (pooled adjusted odds ratio [OR] 1.64; 95% confidence interval [CI], 1.24-2.16, n = 10,853; Figure 1), clinically verified sciatica (pooled adjusted OR 1.35; 95% CI, 1.09-1.68, n = 110,374, Figure 2), and hospitalization or surgery due to a herniated lumbar disc or sciatica (pooled adjusted OR 1.45; 95% CI, 1.16-1.80, n = 337,796; Figure 3). Former smokers had slightly higher risk than never smokers. The pooled adjusted OR was 1.57 (95% CI, 0.98-2.52, n = 8192; Figure 1) for lumbar radicular pain, 1.09 (95% CI, 1.00-1.19, n = 110,374; Figure 2) for clinically verified sciatica, and 1.10 (95% CI, 0.96-1.26, n = 268,630; Figure 3) for hospitalization or surgery due to a herniated lumbar disc or sciatica. The effects of smoking on lumbar radicular pain or clinically verified sciatica did not differ between men (Figure 4) and women (Figure 5).
      Figure thumbnail gr1
      Figure 1A meta-analysis of 9 studies on the effect of smoking on lumbar radicular pain. CI = confidence interval; OR = odds ratio.
      Figure thumbnail gr2
      Figure 2A meta-analysis of 8 studies on the effect of smoking on clinically verified sciatica. CI = confidence interval; OR = odds ratio.
      Figure thumbnail gr3
      Figure 3A meta-analysis of 11 studies on the effect of smoking on hospitalization or surgery due to a herniated lumbar disc or clinically verified sciatica. CI = confidence interval; OR = odds ratio.
      Figure thumbnail gr4
      Figure 4A meta-analysis of 13 studies on the effect of smoking on lumbar radicular pain or clinically verified sciatica among men. CI = confidence interval; OR = odds ratio.
      Figure thumbnail gr5
      Figure 5A meta-analysis of 10 studies on the effect of smoking on lumbar radicular pain or clinically verified sciatica among women. CI = confidence interval; OR = odds ratio.

       Heterogeneity and Publication Bias

      Using meta-regression, heterogeneity was not explained by year of publication, occupational vs nonoccupational population, study design, sex of participants, type of outcome, and methodological quality of included studies. For lumbar radicular pain, high level of heterogeneity was due to 2 studies.
      • Leino-Arjas P.
      • Solovieva S.
      • Kirjonen J.
      • Reunanen A.
      • Riihimäki H.
      Cardiovascular risk factors and low-back pain in a long-term follow-up of industrial employees.
      • Manninen P.
      • Riihimäki H.
      • Heliövaara M.
      Incidence and risk factors of low-back pain in middle-aged farmers.
      I2 dropped from 77.9% to 0% for past smoking (pooled adjusted OR 1.07; 95% CI, 0.94-1.22), and from 78.6% to 19.7% for ever smoking (pooled adjusted OR 1.28; 95% CI, 1.13-1.45) after excluding the 2 studies. For clinically verified sciatica, I2 dropped from 57.5% to 0% for current smoking (pooled adjusted OR 1.40; 95% CI, 1.28-1.54) after excluding one study.
      • Heliövaara M.
      • Mäkelä M.
      • Knekt P.
      • Impivaara O.
      • Aromaa A.
      Determinants of sciatica and low-back pain.
      For hospitalization or surgery due to a herniated lumbar disc or sciatica, I2 dropped from 67.7% to 15.3% for current smoking (pooled adjusted OR 1.28; 95% CI, 1.13-1.43) after excluding one study.
      • An H.S.
      • Silveri C.P.
      • Simpson J.M.
      • et al.
      Comparison of smoking habits between patients with surgically confirmed herniated lumbar and cervical disc disease and controls.
      A funnel plot of 28 studies, 20 on current smoking and 8 on ever smoking, was symmetrical (Supplementary Figure 2, Appendix, available online). P value for Egger test was .56, and the trim-and-fill method imputed only one missing study.

       Sensitivity Analysis

      Sensitivity analyses showed that the effect of smoking on lumbar radicular pain or clinically verified sciatica is independent of study design, study quality, confounding, and publication bias (Table). The effect of current smoking attenuated only slightly after including studies that controlled their estimates for all known confounders, or including only studies that were rated as having low risk of selection bias. Pooled adjusted OR was 1.09 (95% CI, 1.01-1.17) for past smoking and 1.29 (95% CI, 1.14-1.45) for current smoking after limiting the meta-analysis to the studies with low risk of selection bias that controlled also their estimates for all potential confounding factors. Adjustment for publication bias did not change these results.

      Discussion

      This meta-analysis shows that current smoking is a modest risk factor for lumbar radicular pain and clinically verified sciatica. Sensitivity analyses indicate that results are robust across different study designs, sex of participants, and methodological quality of included studies. Thus, the observed association is unlikely due to chance, bias, or confounding.
      The mechanisms by which smoking causes sciatica are only partially known. Smoking reduces perfusion around the intervertebral discs, which can lead to malnutrition of the intervertebral discs.
      • Kauppila L.I.
      • McAlindon T.
      • Evans S.
      • Wilson P.W.
      • Kiel D.
      • Felson D.T.
      Disc degeneration/back pain and calcification of the abdominal aorta. A 25-year follow-up study in Framingham.
      • Shiri R.
      • Viikari-Juntura E.
      • Leino-Arjas P.
      • et al.
      The association between carotid intima-media thickness and sciatica.
      • Korkiakoski A.
      • Niinimäki J.
      • Karppinen J.
      • et al.
      Association of lumbar arterial stenosis with low back symptoms: a cross-sectional study using two-dimensional time-of-flight magnetic resonance angiography.
      Experimental studies found that tobacco smoke is a cause of intervertebral disc degeneration.
      • Wang D.
      • Nasto L.A.
      • Roughley P.
      • et al.
      Spine degeneration in a murine model of chronic human tobacco smokers.
      • Akmal M.
      • Kesani A.
      • Anand B.
      • Singh A.
      • Wiseman M.
      • Goodship A.
      Effect of nicotine on spinal disc cells: a cellular mechanism for disc degeneration.
      • Uematsu Y.
      • Matuzaki H.
      • Iwahashi M.
      Effects of nicotine on the intervertebral disc: an experimental study in rabbits.
      • Oda H.
      • Matsuzaki H.
      • Tokuhashi Y.
      • Wakabayashi K.
      • Uematsu Y.
      • Iwahashi M.
      Degeneration of intervertebral discs due to smoking: experimental assessment in a rat-smoking model.
      Moreover, smoking increases production and release of inflammatory cytokines in intervertebral discs
      • Oda H.
      • Matsuzaki H.
      • Tokuhashi Y.
      • Wakabayashi K.
      • Uematsu Y.
      • Iwahashi M.
      Degeneration of intervertebral discs due to smoking: experimental assessment in a rat-smoking model.
      • Nemoto Y.
      • Matsuzaki H.
      • Tokuhasi Y.
      • et al.
      Histological changes in intervertebral discs after smoking and cessation: experimental study using a rat passive smoking model.
      and interferes with healing process. Among patients with sciatica, smokers not only have a slower healing rate or recovery process than nonsmokers,
      • Haugen A.J.
      • Brox J.I.
      • Grovle L.
      • et al.
      Prognostic factors for non-success in patients with sciatica and disc herniation.
      but they also are at higher risk of recurrent lumbar disc herniation than nonsmokers.
      • Miwa S.
      • Yokogawa A.
      • Kobayashi T.
      • et al.
      Risk factors of recurrent Lumbar disc herniation: a single center study and review of the literature.
      Past smokers are at only a slightly higher risk of sciatica than never smokers. Smoking cessation seems to reduce, but not entirely eliminate, the excess risk. Included studies, however, did not explore the effect of time since stopping smoking on sciatica. In an experimental study in rats,
      • Nemoto Y.
      • Matsuzaki H.
      • Tokuhasi Y.
      • et al.
      Histological changes in intervertebral discs after smoking and cessation: experimental study using a rat passive smoking model.
      the progression of smoking-induced intervertebral disc degeneration ceased and disc degeneration was partially repaired after smoking cessation.
      This meta-analysis has several strengths. There were many large prospective studies on clinically verified sciatica. Several studies confirmed a herniated lumbar disc surgically or by magnetic resonance imaging or computed tomography scan. Furthermore, all studies included in this meta-analysis except 2 controlled their risk estimates for at least age and sex, and many of them controlled their risk estimates for occupation or occupational factors, body mass index, and leisure-time physical activity, as well as other known confounding factors. The observed association between current smoking and sciatica is therefore less likely due to confounding. Further, there was no evidence of publication bias, and pooled estimates were independent of sex of participants, study design, and study quality.
      This meta-analysis also has limitations. Several studies did not differentiate between current and former smokers. A subgroup analysis, however, did not show an impact on pooled estimates for both current and ever smoking. Only a few studies investigated a dose–response relationship between smoking and sciatica. Thus, a meta-analysis of the dose–response relationship between smoking and sciatica was not possible. However, of 6 studies
      • Heliövaara M.
      • Mäkelä M.
      • Knekt P.
      • Impivaara O.
      • Aromaa A.
      Determinants of sciatica and low-back pain.
      • Chiang S.
      • Lin C.
      • Tsai S.
      • et al.
      Cigarette smoking dose as a predictor of need for surgical intervention in patients with lumbar disk herniation.
      • Manninen P.
      • Riihimäki H.
      • Heliövaara M.
      Incidence and risk factors of low-back pain in middle-aged farmers.
      • Jhawar B.S.
      • Fuchs C.S.
      • Colditz G.A.
      • Stampfer M.J.
      Cardiovascular risk factors for physician-diagnosed lumbar disc herniation.
      • Kaila-Kangas L.
      • Leino-Arjas P.
      • Riihimäki H.
      • Luukkonen R.
      • Kirjonen J.
      Smoking and overweight as predictors of hospitalization for back disorders.
      that explored a dose–response relation between smoking and lumbar radicular pain, or clinically verified sciatica, 4 studies
      • Chiang S.
      • Lin C.
      • Tsai S.
      • et al.
      Cigarette smoking dose as a predictor of need for surgical intervention in patients with lumbar disk herniation.
      • Miranda H.
      • Viikari-Juntura E.
      • Martikainen R.
      • Takala E.P.
      • Riihimäki H.
      Individual factors, occupational loading, and physical exercise as predictors of sciatic pain.
      • Jhawar B.S.
      • Fuchs C.S.
      • Colditz G.A.
      • Stampfer M.J.
      Cardiovascular risk factors for physician-diagnosed lumbar disc herniation.
      • Kaila-Kangas L.
      • Leino-Arjas P.
      • Riihimäki H.
      • Luukkonen R.
      • Kirjonen J.
      Smoking and overweight as predictors of hospitalization for back disorders.
      showed a dose–response relation between pack-years and lumbar radicular pain,
      • Miranda H.
      • Viikari-Juntura E.
      • Martikainen R.
      • Takala E.P.
      • Riihimäki H.
      Individual factors, occupational loading, and physical exercise as predictors of sciatic pain.
      and between pack-years,
      • Chiang S.
      • Lin C.
      • Tsai S.
      • et al.
      Cigarette smoking dose as a predictor of need for surgical intervention in patients with lumbar disk herniation.
      • Kaila-Kangas L.
      • Leino-Arjas P.
      • Riihimäki H.
      • Luukkonen R.
      • Kirjonen J.
      Smoking and overweight as predictors of hospitalization for back disorders.
      number of years smoked,
      • Chiang S.
      • Lin C.
      • Tsai S.
      • et al.
      Cigarette smoking dose as a predictor of need for surgical intervention in patients with lumbar disk herniation.
      or number of cigarettes smoked per day,
      • Jhawar B.S.
      • Fuchs C.S.
      • Colditz G.A.
      • Stampfer M.J.
      Cardiovascular risk factors for physician-diagnosed lumbar disc herniation.
      and clinically verified sciatica. In addition, the included studies did not investigate a relationship between time since smoking cessation and sciatica.
      In summary, the current meta-analysis suggests that smoking is a risk factor for lumbar radicular pain and clinically verified sciatica. The harmful effect of smoking on lumbar intervertebral discs is modest, and stopping smoking appears to reduce the excess risk.

      Appendix

      Supplementary Table 1PubMed Search Strategy Made on March 19, 2015
      SearchQueryNo of Items Found
      #1Smoking[Mesh] OR smoking[Text Word] OR tobacco[Mesh] OR tobacco[Text Word] OR tobacco products[Mesh] OR cigarette[Text Word] OR cigar[Text Word] OR pipe[Text Word] OR smoke[Mesh] OR smoke[Text Word] OR smok*[Text Word] OR tobacco use disorder[Mesh] OR nicotine[Mesh] OR nicotine[Text Word] OR life style[Mesh] OR life style[Text Word] OR risk factors[Mesh] OR risk factor[Text Word]934,614
      #2Sciatica[Mesh] OR sciatica[Text Word] OR intervertebral disk displacement[Mesh] OR intervertebral disk displacement[Text Word] OR lumbar radicular pain[Text Word] OR sciatic pain[Text Word] OR sciatic neuropathy[Mesh] OR sciatic neuropathy[Text Word] OR sciatic syndrome[Text Word] OR lumbosciatic syndrome[Text Word] OR lumbosacral radicular syndrome[Text Word] OR disc herniation[Text Word] OR herniated lumbar disc[Text Word] OR prolapsed lumbar disc[Text Word] OR disc protrusion[Text Word] OR herniated nucleus pulposus[Text Word] OR spinal diseases[Mesh] OR spinal diseases[Text Word] OR back pain[Mesh] OR back pain[Text Word] OR back disorders[Text Word]135,150
      Final#1 AND #26006
      Supplementary Table 2Quality Assessment
      Type of BiasCriteria DefinitionClassification (Potential for Bias)
      Selection biasSampling method of the study population, representativeness (response rate, difference between responders and nonresponders, investigate and control of variables in case of difference between responders and nonresponders)Low: Target population defined as representative of the general population or subgroup of the general population (specific age group, women, men, specific geographic area, and specific occupational group) and response rate is 80% or more.

      Moderate: Target population defined as somewhat representative of the general population, a restricted subgroup of the general population, response rate 60%-79%.

      High: Target population defined as “self-referred”/ volunteers, or response rate <60%.
      Performance biasValid and reliable assessment of exposure

      Exposure assessors blinded to outcome status
      Low: Current smokers compared with never smokers, and former smokers compared with never smokers. Moderate: Current smokers compared with never or former smokers, or current and former smokers combined in a single group.
      Detection biasClear definition of outcome

      Standard method for outcome assessment

      Outcome assessors blinded to exposure status
      Low: Clinical diagnosis of sciatica.

      Moderate: Self-reported physician-diagnosed sciatica or using a validated questionnaire.
      ConfoundingMatching 2 groups

      Stratification

      Statistical analysis
      Low: Controlled for most of confounding factors.

      Moderate: Controlled for few confounding factors, particularly age and sex.

      High: Not controlled for both age and sex, or controlled for two or fewer confounding factors.
      Attrition biasWithdrawals and dropout rates

      Size of missing data
      Low: Follow up participation rate of 80% or higher, or missing data of <20%.

      Moderate: Follow-up participation rate of 60%-79%, or missing data of 20%-40%.

      High: Follow-up participation rate of <60%, or missing data of more than 40%.
      Supplementary Table 3Studies Included in the Meta-analysis
      First Author and Year of PublicationCountryStudy PopulationAge Range (Years)SexSample Size (in Analysis)SmokingOutcomeQuality Assessment: Risk of bias
      Low means low risk of bias and high means high risk of bias.
      ResultsAdjustment for Other Covariates
      SelectionPerformanceDetectionConfoundingAttrition
      Cross-sectional studies
       Karjalainen, 2013
      • Violante F.S.
      • Fiori M.
      • Fiorentini C.
      • et al.
      Associations of psychosocial and individual factors with three different categories of back disorder among nursing staff.
      FinlandAdolescent18Both1987Smokers vs never smokers, pack-yearsSciatic pain during the past 12 months with no consultation or pain medication (mild sciatica), sciatic pain required pain medication or consultation with a physician (severe sciatica)ModerateLowModerateLowModerateOR of mild or severe sciatica 0.93 (CI, 0.57-1.52)Sex, BMI, physical activity, sedentary behavior, physical workload and low back pain at age 16 y
       Korovessis, 2012
      • Schumann B.
      • Bolm-Audorff U.
      • Bergmann A.
      • et al.
      Lifestyle factors and lumbar disc disease: results of a German multi-center case-control study (EPILIFT).
      GreeceGeneral population20 years or older. 60.5% were younger than 46Both674 (254 men and 420 women)Smokers vs nonsmokersLow back pain radiating to the posterior surface of the leg during the past 6 monthsModerateModerateModerateModerateLow31.7% of smokers and 20.9% of nonsmokers had sciatic pain.

      Estimated unadjusted prevalence ratio 1.515 (CI, 1.16-1.97).

      Adjusted OR 2.1 (CI, 1.4-3.1)
      Age, sex, marital status, and work status
       Leino-Arjas, 2008
      • Zhang Y.G.
      • Sun Z.
      • Zhang Z.
      • Liu J.
      • Guo X.
      Risk factors for lumbar intervertebral disc herniation in Chinese population: a case-control study.
      FinlandGeneral population30 years or olderBoth5644 (2574 men and 3070 women)Never smokers; past smokers; current smokers, pack-years ≤20; and current smokers, pack-years >20Physician diagnosed sciaticaLowLowLowModerateLowEstimated unadjusted prevalence ratio for past smoking 1.71 (CI, 0.83-3.51) for men, 1.08 (CI, 0.62-1.89) for women, and 1.39 (CI, 0.92-2.10) for men and women combined.

      Estimated unadjusted prevalence ratio for current smoking 2.08 (CI, 1.01-4.30) for men, 1.04 (CI, 0.55-1.98) for women, and 1.59 (CI, 1.03-2.44) for men and women combined.

      Estimated unadjusted prevalence ratio for ever smoking 1.867 (CI, 0.94-3.71) for men, 1.064 (CI, 0.64-1.75) for women, and 1.34 (CI, 0.97-1.85) for men and women combined.

      Adjusted OR for past smoking 1.58 (CI, 0.78-3.23) for men, 1.12 (CI, 0.64-1.96) for women, and 1.28 (CI, 0.82-1.98) for men and women combined.

      Adjusted OR for current smoking 2.29 (CI, 1.26-4.15) for men, 1.22 (CI, 0.63-2.37) for women, and 1.73 (CI, 1.11-2.69) for men and women combined.

      Adjusted OR for ever smoking 1.965 (CI, 1.24-3.10) for men, 1.16 (CI, 0.757-1.780) for women, and 1.484 (CI, 1.086-2.028) for men and women combined.
      Age and sex
       Younes, 2006
      • Heliövaara M.
      • Knekt P.
      • Aromaa A.
      Incidence and risk factors of herniated lumbar intervertebral disc or sciatica leading to hospitalization.
      TunisiaGeneral populationAge range 15-99, mean age 36.1 ± 16.7Both4380 (2221 men and 2159 women)Smokers vs nonsmokersPhysician diagnosed sciatica during past 12 monthsLowModerateLowHighLowThe prevalence of smoking was 47.4% in those with sciatica and 18.5% in those free from sciatica.

      Estimated prevalence ratio 2.56 (CI, 2.05-3.18)
      Unadjusted
       Violante, 2004
      • Sørensen I.G.
      • Jacobsen P.
      • Gyntelberg F.
      • Suadicani P.
      Occupational and other predictors of herniated lumbar disc disease-a 33-year follow-up in the Copenhagen male study.
      ItalyNursesMean age 35.7 ± 8.5Women858Current or past smokers vs never smokersSelf-reported instrumentally diagnosed herniation of a lumbar discModerateModerateModerateLowLowEstimated unadjusted prevalence ratio 1.034 (CI, 0.745-1.437).

      Adjusted OR 0.95 (CI, 0.49-1.85)
      Age, BMI, motherhood, scoliosis, trauma or fractures of spine or pelvis, sport, work environment / job satisfaction, other diagnosed spine pathologies, coping strategies, stress related psychosomatic symptoms, and depression-related tendencies
       Kostova, 2001
      • Wahlström J.
      • Burstrom L.
      • Nilsson T.
      • Järvholm B.
      Risk factors for hospitalization due to lumbar disc disease.
      BulgariaWorkers of a fertilizer plantNot reportedBoth895Current smokers vs never or past smokers. Worker who had smoked more than 20 years compared with those smoked less than 20 years and daily consumption of more than 20 cigarettes compared with 20 or less.Clinically defined lumbosacral radicular syndromeModerateModerateLowHighLowEstimated prevalence ratio 0.63 (CI, 0.44-0.92) for current smokers vs never or past smokers.

      Estimated prevalence ratio 1.87 (CI, 1.20-2.91) for individuals who had smoked for more than 20 years vs smoked <20 years
      Unadjusted
       Heliövaara, 1991
      • Kaila-Kangas L.
      • Leino-Arjas P.
      • Riihimäki H.
      • Luukkonen R.
      • Kirjonen J.
      Smoking and overweight as predictors of hospitalization for back disorders.
      FinlandGeneral population30-64Both5673 (2727 men and 2946 women)Never, past, or current smokers, and number of cigarettes smoked per dayPhysician diagnosed sciaticaLowLowLowLowLowEstimated unadjusted prevalence ratio 1.09 (CI, 0.82-1.45) for past smoking, 1.01 (CI, 0.73-1.40) for cigars, pipes or 1-19 cigarettes per day, and 1.24 (CI, 0.87-1.77) for 20 or more cigarettes per day.

      Estimated unadjusted prevalence ratio 1.11 (CI, 0.87-1.41) for current smoking and 1.10 (CI, 0.92-1.32) for ever smoking.

      Adjusted OR 0.9 (CI, 0.6-1.3) for past smoking, 0.9 (CI, 0.6-1.3) for cigars, pipes or 1-19 cigarettes/day and 1.1 (CI, 0.7-1.6) for 20 or more cigarettes per day.

      Estimated adjusted OR 0.988 (CI, 0.745-1.311) for current smoking and 0.957 (CI, 0.762-1.202) for ever smoking
      Age, sex, body height, prior traumatic back injury, physical work load, driving motor vehicles, occupational mental stress, body mass index and alcohol consumption
      Case control studies
       Chiang, 2014
      • Saftic R.
      • Grgic M.
      • Ebling B.
      • Splavski B.
      Case-control study of risk factors for lumbar intervertebral disc herniation in Croatian island populations.
      TaiwanCases were patients with lumbar disc herniation and controls were a random sample of other inpatientsMean age 39 ± 18.4Both, 69% were men822 (391 cases and 431 controls). Of cases 264 had lumbar discectomyCurrent or past smokers vs never smokers. Number of cigarettes smoked per day, number of years smoked, and pack-yearsHospitalization due to lumbar disc herniation confirmed by CT or MRI.

      Surgery due to lumbar intervertebral disc herniation
      ModerateModerateLowModerateLowUnadjusted 1.82 (CI, 1.37-2.43) for hospitalization, and 1.63 (1.21-2.21) for surgery.

      Adjusted OR 1.51 (CI, 1.09-2.10) for hospitalization, and 1.76 (1.26-2.47) for surgery.

      Estimated unadjusted OR 1.73 (CI, 1.25-2.38) and estimated adjusted OR 1.63 (CI, 1.13-2.34) for hospitalization or surgery.

      A dose-response relation between pack-years and number of years smoked, and both outcomes
      Sex, age, height, body weight, and BMI
       Zhang, 2013
      • An H.S.
      • Silveri C.P.
      • Simpson J.M.
      • et al.
      Comparison of smoking habits between patients with surgically confirmed herniated lumbar and cervical disc disease and controls.
      ChinaCases were patients with lumbar disc herniation and controls were a random sample of other inpatients or participants of medical examinationsMean age 46.2 ± 13.1 for cases and 47.4 ± 12.9 for controlsBoth260 (128 cases and 132 controls)Current or past smokers vs never smokersLumbar disc herniation confirmed by CT or MRIModerateModerateLowModerateLowOR 1.286 (CI, 0.770-2.145)Matched by age, sex, race, and living area
       Schumann, 2010
      • Rivinoja A.E.
      • Paananen M.V.
      • Taimela S.P.
      • et al.
      Sports, smoking, and overweight during adolescence as predictors of sciatica in adulthood: a 28-year follow-up study of a birth cohort.
      GermanyPatients with lumbar disc herniation and population controls25-70Both1458 (561 patients and 897 controls). 736 men and 722 womenCurrent or past smokers vs never smokers, pack-yearsIn- or outpatients who received treatment due to lumbar disc herniation (confirmed by CT or MRI) with radiculopathy and with sensory or motor deficitsHighModerateLowModerateLowEstimated unadjusted OR 1.33 (CI, 0.98-1.81) for men, 1.27 (CI, 0.94-1.72) for women, and 1.30 (CI, 1.05-1.61) for men and women combined.

      Adjusted OR 1.18 (CI, 0.92-1.52) for men, 1.16 (CI, 0.90-1.50) for women, and 1.17 (CI, 0.98-1.40) for men women combined
      Sex-specific adjusted for age, region, and cumulative physical workload
       Zhang, 2009
      • Mattila V.M.
      • Saarni L.
      • Parkkari J.
      • Koivusilta L.
      • Rimpelä A.
      Early risk factors for lumbar discectomy: an 11-year follow-up of 57,408 adolescents.
      ChinaPatients hospitalized due to lumbar disc disorders, and controls a random sample of inpatients or participants of medical examinationMean age 45 for cases and 47 for controlsBoth4180 (2010 cases and 2170 controls)Current or past smokers vs never smokersHospitalization due to lumbar disc disordersModerateModerateLowModerateLow41.1% of cases and 31.2% of controls were smokers.

      Estimated OR 1.54 (CI, 1.36-1.75)
      Matched by age, sex, race, and living area
       Saftic, 2006
      • Kauppila L.I.
      • McAlindon T.
      • Evans S.
      • Wilson P.W.
      • Kiel D.
      • Felson D.T.
      Disc degeneration/back pain and calcification of the abdominal aorta. A 25-year follow-up study in Framingham.
      CroatiaA random sample from nine villages. Cases were patients with lumbar discectomy due to lumbar intervertebral disc herniation, and controls were a random sample of healthy individuals18 or olderBoth335 (67 cases and 268 controls)Current or past smokers vs never smokersSurgery due to lumbar intervertebral disc herniationLowModerateLowModerateLowEstimated OR 0.99 (CI, 0.57-1.69) for ever smokingMatched by age, sex, and village of residence or immigrant status
       An, 1994
      • Shiri R.
      • Viikari-Juntura E.
      • Leino-Arjas P.
      • et al.
      The association between carotid intima-media thickness and sciatica.
      USACases were patients with lumbar discectomy due to disc herniation, and controls were other medical or surgical inpatientsAge range 16-78, mean age 44.5 for both groupsBoth368 (163 cases and 205 controls). 228 men and 140 womenNever, past, or current smokersLumbar discectomy due to intervertebral disc herniationModerateLowLowModerateLowEstimated OR 1.25 (CI, 0.70-2.23) for past smoking, 3.14 (CI, 1.92-5.13) for current smoking, and 2.21 (CI, 1.45-3.36) for ever smoking.

      Estimated OR for ever smoking 3.41 (CI, 1.97-5.91) for men and 1.14 (CI, 0.58-2.23) for women
      Matched by age and sex
       Heliövaara, 1987
      • Korkiakoski A.
      • Niinimäki J.
      • Karppinen J.
      • et al.
      Association of lumbar arterial stenosis with low back symptoms: a cross-sectional study using two-dimensional time-of-flight magnetic resonance angiography.
      FinlandGeneral population20-59Both2742 (592 cases, 2150 controls). 1672 men and 1070 womenCurrent smokers vs nonsmokersHospitalization due to herniated lumbar disc or sciaticaLowModerateLowLowLowEstimated unadjusted OR of herniated disc 1.36 (CI, 1.00-1.85) and multivariate OR 1.3 (CI, not reported) for men.

      Estimated unadjusted OR of herniated disc or sciatica 1.16 (CI, 0.92-1.47) and multivariate OR 1.0 (CI, not reported) for men.

      Estimated unadjusted OR of herniated disc 1.175 (CI, 0.707-1.954) and multivariate OR 1.0 (CI, not reported) for women.

      Estimated unadjusted OR of herniated disc or sciatica 1.372 (CI, 0.947-1.987) and multivariate OR 1.4 (CI, not reported) for women.

      Estimated unadjusted OR of herniated disc 1.269 (CI, 0.994-1.620) for men and women combined.

      Estimated unadjusted OR of herniated disc or sciatica 1.197 (CI, 0.995-1.441) for men and women combined.

      Estimated adjusted OR of herniated disc 1.212 (CI, 0.931-1.576) for men and women combined.

      Estimated adjusted OR of herniated disc or sciatica 1.10 (CI, 0.90-1.34) for men and women combined
      Cases were matched with controls for age, sex and place of residence.

      Adjusted for occupational group, social class, marital status, number of birth (for women), leisure time physical activity, chronic cough, number of distress symptoms, use of medication, and frequent use of analgesics
       Kelsey, 1984
      • Wang D.
      • Nasto L.A.
      • Roughley P.
      • et al.
      Spine degeneration in a murine model of chronic human tobacco smokers.
      USACases were in- or outpatients with prolapsed lumbar intervertebral disc, or sciatica. Controls were patients admitted to the same hospitals or practices for conditions not related to the spineAge range 20-64Both650 (325 cases and 325 controls)Never, past, or current smokersLumbar disc herniation (165 of 325 cases), or sciaticaModerateLowLowModerateLowUnadjusted OR 1.0 (CI, 0.6-1.7) for past smoking, and 1.7 (CI, 1.0-2.5) for current smoking.

      Estimated unadjusted OR 1.35 (CI, 0.95-1.90) for ever smoking.

      Adjusted OR 1.2 (CI, 1.0-1.4) for current smoking for 10-cigarette increase
      74% of cases were matched with controls for age and sex.

      OR adjusted for driving, lifting objects >11.3 kg >25 times per day while twisting body, knees not bent, and lifting objects >11.3 kg >25 times per day while twisting body, knees bent
      Cohort studies
       Sørensen 2011
      • Akmal M.
      • Kesani A.
      • Anand B.
      • Singh A.
      • Wiseman M.
      • Goodship A.
      Effect of nicotine on spinal disc cells: a cellular mechanism for disc degeneration.
      DenmarkOccupational population (14 private or public companies)

      A 33-y follow-up
      40-59 at baselineMen3831Never, past, or current smokersHospitalization due to herniated lumbar disc diseaseLowLowLowModerateLowEstimated unadjusted RR 1.32 (CI, 0.52-3.31) for past smoking, 0.80 (0.34-1.87) for current smoking, and 0.91 (CI, 0.39-2.08) for ever smoking.

      Estimated age-adjusted RR 1.43 (CI, 0.56-3.63) for past smoking, 0.95 (CI, 0.40-2.23) for current smoking and 1.146 (CI, 0.61-2.16) for ever smoking
      Age
       Matsudaira, 2013
      • Uematsu Y.
      • Matuzaki H.
      • Iwahashi M.
      Effects of nicotine on the intervertebral disc: an experimental study in rabbits.
      JapanOccupational population (16 offices of participating organizations).

      A 2-year follow-up
      <40 y 37.6%, 40-49 y. 29.6%, and 50 y or older 32.8%Both, 88.5% were men765Heavy smokers (Brinkmann smoking index 400 or higher) vs nonheavy smokersLow back pain radiating blow the knee during the follow-upModerateModerateModerateModerateModerateUnadjusted OR 1.35 (CI, 0.89-2.03).

      Adjusted OR 1.20 (CI, 0.76-1.88)
      Age and sex
       Wahlström, 2012
      • Oda H.
      • Matsuzaki H.
      • Tokuhashi Y.
      • Wakabayashi K.
      • Uematsu Y.
      • Iwahashi M.
      Degeneration of intervertebral discs due to smoking: experimental assessment in a rat-smoking model.
      SwedenConstruction workers.

      A 17-year follow-up
      20-65Men263,529Never, past, or current smokersHospitalization due to lumbar disc diseaseLowLowLowLowModerateAdjusted RR 1.08 (CI, 0.94-1.25) for past smoking and 1.27 (CI, 1.15-1.39) for current smoking, and 1.208 (CI, 1.117-1.308) for ever smokingAge, occupation, body height, body weight and time period of the diagnosis
       Kääriä, 2011
      • Nemoto Y.
      • Matsuzaki H.
      • Tokuhasi Y.
      • et al.
      Histological changes in intervertebral discs after smoking and cessation: experimental study using a rat passive smoking model.
      FinlandMunicipal employees.

      5 to 7 years follow-up
      40-60 at baselineBoth5261 (1098 men and 4163 women)Never, past, or current smokersLow back pain radiating to the calf or the foot during the follow-upModerateLowModerateLowLowEstimated unadjusted RR for past smoking 1.24 (CI, 0.79-1.94) for men and 2.34 (CI, 1.45-3.77) for women.

      Estimated unadjusted RR for current smoking 2.01 (CI, 1.32-3.06) for men and 3.50 (CI, 2.23-5.50) for women.

      Estimated unadjusted RR for ever smoking 1.58 (CI, 1.08-2.31) for men and 2.87 (CI, 1.91-4.32) for women.

      Estimated unadjusted RR 1.670 (CI, 1.204-2.317) for past smoking, 2.601 (CI, 1.912-3.537) for current smoking, and 2.113 (CI, 1.689-2.644) for ever smoking for men and women combined.

      Adjusted OR for past smoking 1.00 (CI, 0.73-1.36) for men and 1.06 (CI, 0.90–1.24) for women.

      Adjusted OR for current smoking 1.36 (CI, 0.97-1.89) for men and 1.45 (CI, 1.23–1.71) for women.

      Estimated adjusted OR for ever smoking 1.154 (CI, 0.919-1.449) for men and 1.234 (CI, 1.100-1.385) for women.

      Estimated adjusted OR 1.047 (CI, 0.908-1.207) for past smoking, 1.432 (CI, 1.235-1.660) for current smoking, and 1.218 (CI, 1.099-1.349) for ever smoking for men and women combined
      Age, occupational class, BMI, leisure-time physical activity, pain-related factors, and physical work factors (for men only)
       Rivinoja, 2011
      • Haugen A.J.
      • Brox J.I.
      • Grovle L.
      • et al.
      Prognostic factors for non-success in patients with sciatica and disc herniation.
      FinlandAdolescent

      A 28-year follow-up
      14 at baseline 42 at the end of follow-upBoth9016 (4535 males and 4481 females)Occasional or daily smokers vs never smokers or those smoked less than a few timesLumbar discectomy due to sciaticaLowLowLowLowModerateAdjusted OR 1.35 (CI, 0.88-2.07) for males, and 1.0 (CI, 0.38-2.60) for females.

      Estimated adjusted OR 1.285 (CI, 0.869-1.899) for males and females combined
      Participation in sports, participation in sports with risk of injury, membership in a sport club, and overweight
       Mattila, 2008
      • Miwa S.
      • Yokogawa A.
      • Kobayashi T.
      • et al.
      Risk factors of recurrent Lumbar disc herniation: a single center study and review of the literature.
      FinlandAdolescent

      A 11.3-year follow-up
      14, 16 or 18 at baselineBoth57,408 (26,688 boys and 30,719 girls)Smoking daily vs not dailyLumbar discectomyModerateLowLowModerateLowAge-adjusted HR 1.6 (CI, 1.2-2.2) for males and 0.7 (CI, 0.4-1.2) for females.

      Estimated age-adjusted HR 1.319 (CI, 1.012-1.720) for males and females combined.

      HR 1.5 (CI, 1.1-2.2) for males adjusted for age and timing of puberty.

      Estimated adjusted HR 1.207 (CI, 0.901-1.619) for males and females combined (a rough estimate)
      Adjusted for age for females and for age and timing of puberty for males
       Jhawar, 2006
      • Kanayama M.
      • Togawa D.
      • Takahashi C.
      • Terai T.
      • Hashimoto T.
      Cross-sectional magnetic resonance imaging study of lumbar disc degeneration in 200 healthy individuals.
      USANurses.

      A 16-year follow-up
      30-55 at baselineWomen98,407Never, past, or current smokers, number of cigarettes smoked per daySelf-reported ever physician diagnosed herniated lumbar disc confirmed by CT scan or MRILowLowModerateLowLowAge-adjusted RR 1.13 (CI, 1.03-1.23) for past smoking, 1.28 (CI, 1.16-1.40) for current smoking, and 1.20 (CI, 1.12-1.28) for ever smoking.

      Multivariate RR 1.10 (CI, 1.00-1.20) for past smoking, 1.38 (CI, 1.26-1.52) for current smoking, and 1.23 (CI, 1.15-1.31) for ever smoking.

      A dose-response relation between number of cigarettes smoked per day and lumbar disc herniation
      Age, BMI, physical exercise, employment status, high cholesterol, hypertension, diabetes, physician visitation within the last year, family history of myocardial infarction before age 60
       Leino-Arjas, 2006
      • Mitchell M.D.
      • Mannino D.M.
      • Steinke D.T.
      • Kryscio R.J.
      • Bush H.M.
      • Crofford L.J.
      Association of smoking and chronic pain syndromes in Kentucky women.
      FinlandMetal industry employees.

      A 28-year follow-up
      18-64Both530 (345 men and 185 women)Never, past, or current smokersFrequent radiating low back pain during the past 12 monthsLowLowModerateModerateModerateEstimated unadjusted RR for past smoking 2.41 (CI, 1.45-4.02) for men, 1.49 (CI, 0.81-2.72) for women, and 1.76 (CI, 1.23-2.52) for both sexes.

      Estimated unadjusted RR for current smoking 2.22 (CI, 1.30-3.77) for men, 2.19 (CI, 1.40-3.43) for women, and 1.92 (CI, 1.35-2.72) for both sexes.

      Estimated unadjusted RR for ever smoking 2.32 (CI, 1.45-3.72) for men, 1.88 (CI, 1.24-2.86) for women, and 1.84 (CI, 1.35-2.50) for both sexes.

      Adjusted OR for past smoking 2.85 (CI, 1.45-5.58) for men, 1.78 (CI, 0.63-5.01) for women, and 2.48 (CI, 1.41-4.36) for men and women combined.

      Adjusted OR for current smoking 2.12 (CI, 1.06-4.24) for men, 2.71 (CI, 1.13-6.49) for women, and 2.33 (CI, 1.35-4.01) for men and women combined.

      Adjusted OR for ever smoking 2.47 (CI, 1.52-4.00) for men, 2.28 (CI, 1.17-4.44) for women, and 2.40 (CI, 1.62-3.55) for men and women combined
      Age, sex and occupational class
       Kaila-Kangas, 2003
      • Södervall J.
      • Karppinen J.
      • Puolitaival J.
      • et al.
      Heart rate variability in sciatica patients referred to spine surgery: a case control study.
      FinlandMetal industry employees.

      A 28-year follow-up
      18-64Both902Never, past, or current smokers, pack-yearsHospitalization due to intervertebral disc disordersLowLowLowLowLowEstimated unadjusted RR 1.32 (CI, 0.48-3.57) for past smoking, 2.36 (CI, 1.12-4.96) for current smoking, and 1.99 (CI, 0.97-4.10) for ever smoking.

      Adjusted RR 1.64 (CI, 0.57-4.76) for past smoking, 2.30 (CI, 0.91-5.82) for current smoking 1-9 pack-y, and 3.41 (CI, 1.29-8.98) for current smoking >9 pack-y

      Estimated adjusted RR 2.776 (CI, 1.420-5.428) for current smoking, and 2.389 (CI, 1.355-4.211) for ever smoking.

      A dose-response relation between pack-years and hospitalization due to disc disorders
      Age, sex, occupational class, strenuous physical work, exercise, BMI, chronic back disorders and stress syndrome
       Miranda, 2002
      • Gökmen F.
      • Zateri C.
      • Akman T.
      • et al.
      Relationship between the localization of the disc herniation and obesity in patients with symptomatic lumbar disc herniation.
      FinlandForest industry workers.

      A 1-year follow-up
      Mean age 45Both2045Never, past, or current smokers. Pack-years.Radiating low back pain longer than 7 days during the past 12 monthsHighLowModerateLowLowEstimated unadjusted RR 1.33 (CI, 1.10-1.61) for past smoking, 1.51 (CI, 1.24-1.84) for current smoking, and 1.28 (CI, 1.14-1.43) for ever smoking.

      Adjusted OR 1.3 (CI, 0.9-2.0) for past smoking, 1.95 (CI, 1.37-2.78) for current smoking, and 1.64 (CI, 1.25-2.13) for ever smoking.

      A dose-response relation between pack-years and sciatic pain
      Age, sex, mental stress, walking, twisting movements of the trunk, working in kneeling or squatting position, and working with hand above shoulder level
       Manninen, 1995
      • Rasmussen C.
      Lumbar disc herniation: favourable outcome associated with intake of wine.
      FinlandFarmers.

      A 12-year follow-up
      Age range 45-54 and mean age 49.3 ± 2.8 at baselineMen193Never, past, or current smokers. Pack-years.Low back pain radiating to the foot during past 12 monthsHighLowModerateLowModerateUnadjusted RR 3.74 (CI, 1.21-11.54) for past smoking and 2.96 (CI, 0.91-9.64) for current smoking, and 3.35 (CI, 1.16-9.66) for ever smoking.

      Adjusted OR 13.1 (CI, 2.66-65.1) for past smoking and 9.6 (CI, 1.73-53.0) for current smoking, and 11.33 (CI, 3.52-36.45) for ever smoking.

      No dose-response relation between pack-years and sciatic pain
      Age, height, mental stress, BMI, farm production, occupational class and pain in the joints
       Riihimäki, 1994
      • Graver V.
      • Haaland A.K.
      • Magnaes B.
      • Loeb M.
      Seven-year clinical follow-up after lumbar disc surgery: results and predictors of outcome.
      FinlandMachine operators, carpenters, and office workers.

      A 3-year follow-up
      25-49Men1149Current or past smokers vs never smokers.

      Smoking status was inquired at the follow-up
      Low back pain radiating to the leg. 3-year cumulative incidence of sciatic painLowModerateModerateModerateLowOR 1.29 (CI, 0.98-1.69)Occupation, physical exercise, and history of other low back pain
       Riihimäki, 1989
      • Suk K.S.
      • Lee H.M.
      • Moon S.H.
      • Kim N.H.
      Recurrent lumbar disc herniation: results of operative management.
      FinlandConcrete reinforcement workers or house painters.

      A 5-year follow-up
      25-54Men163Never, past, or current smokers5-year cumulative incidence of sciatic painModerateLowModerateModerateModerateUnadjusted RR 1.0 (CI, 0.4-2.3) for past smoking and 1.0 (CI, 0.4-2.2) for current smoking.

      Age-adjusted RR 1.0 (CI, 0.4-2.1) for past smoking and 1.0 (CI, 0.5-2.0) for current smoking.

      Estimated unadjusted RR 1.0 (CI, 0.54-1.84) for ever smoking.

      Estimated age-adjusted RR 1.0 (CI, 0.59-1.70) for ever smoking
      Age
      BMI = body mass index; CI = 95% confidence interval; CT = computed tomography; HR = hazard ratio; MRI = magnetic resonance imaging; OR = odds ratio; RR = relative risk.
      Low means low risk of bias and high means high risk of bias.
      Figure thumbnail fx1
      Supplementary Figure 1Flow chart of the search strategy and selection of studies.
      Figure thumbnail fx2
      Supplementary Figure 2A funnel plot of 28 studies on the association between current smoking (20 studies) or ever smoking (8 studies), and lumber radicular pain or clinically verified sciatica. P-value for Egger test = .56.

      References

        • Koes B.W.
        • van Tulder M.W.
        • Peul W.C.
        Diagnosis and treatment of sciatica.
        BMJ. 2007; 334: 1313-1317
        • Younes M.
        • Bejia I.
        • Aguir Z.
        • et al.
        Prevalence and risk factors of disk-related sciatica in an urban population in Tunisia.
        Joint Bone Spine. 2006; 73: 538-542
        • Leino-Arjas P.
        • Kauppila L.
        • Kaila-Kangas L.
        • Shiri R.
        • Heistaro S.
        • Heliövaara M.
        Serum lipids in relation to sciatica among Finns.
        Atherosclerosis. 2008; 197: 43-49
        • Heliövaara M.
        • Mäkelä M.
        • Knekt P.
        • Impivaara O.
        • Aromaa A.
        Determinants of sciatica and low-back pain.
        Spine (Phila Pa 1976). 1991; 16: 608-614
        • Shiri R.
        • Lallukka T.
        • Karppinen J.
        • Viikari-Juntura E.
        Obesity as a risk factor for sciatica: a meta-analysis.
        Am J Epidemiol. 2014; 179: 929-937
        • Shiri R.
        • Karppinen J.
        • Leino-Arjas P.
        • et al.
        Cardiovascular and lifestyle risk factors in lumbar radicular pain or clinically defined sciatica: a systematic review.
        Eur Spine J. 2007; 16: 2043-2054
        • Shiri R.
        • Karppinen J.
        • Leino-Arjas P.
        • Solovieva S.
        • Viikari-Juntura E.
        The association between smoking and low back pain: a meta-analysis.
        Am J Med. 2010; 123: 87.e7-87.e35
        • Moher D.
        • Liberati A.
        • Tetzlaff J.
        • Altman D.G.
        • Group P.
        Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement.
        Ann Intern Med. 2009; 151 (W64): 264-269
        • Shiri R.
        • Karppinen J.
        • Leino-Arjas P.
        • Solovieva S.
        • Viikari-Juntura E.
        The association between obesity and low back pain: a meta-analysis.
        Am J Epidemiol. 2010; 171: 135-154
        • Armijo-Olivo S.
        • Stiles C.R.
        • Hagen N.A.
        • Biondo P.D.
        • Cummings G.G.
        Assessment of study quality for systematic reviews: a comparison of the Cochrane Collaboration Risk of Bias Tool and the Effective Public Health Practice Project Quality Assessment Tool: methodological research.
        J Eval Clin Pract. 2012; 18: 12-18
        • Woolf B.
        On estimating the relation between blood group disease.
        Ann Hum Genet. 1955; 19: 251-253
      1. Higgins J, Green S. Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0. The Cochrane Collaboration; 2009 [updated March 2011]. Available at: www.cochrane-handbook.org. Accessed July 2015.

        • Chiang S.
        • Lin C.
        • Tsai S.
        • et al.
        Cigarette smoking dose as a predictor of need for surgical intervention in patients with lumbar disk herniation.
        J Med Sci. 2014; 34: 23-29
        • Borenstein M.
        • Hedges L.
        • Higgins J.
        • Rothstein H.
        Introduction to Meta-analysis.
        John Wiley & Sons, Ltd., New York2009
        • Higgins J.P.
        • Thompson S.G.
        Quantifying heterogeneity in a meta-analysis.
        Stat Med. 2002; 21: 1539-1558
        • Ioannidis J.P.
        • Patsopoulos N.A.
        • Evangelou E.
        Uncertainty in heterogeneity estimates in meta-analyses.
        BMJ. 2007; 335: 914-916
        • Morton S.C.
        • Adams J.L.
        • Suttorp M.J.
        • Shekelle P.G.
        Meta-regression Approaches: What, Why, When, and How? Technical Review 8.
        Southern California–RAND Evidence-based Practice Center, Santa Monica, CA2004
        • Rothstein H.
        • Sutton A.
        • Borenstein M.
        Publication Bias in Meta-analysis: Prevention, Assessment and Adjustments.
        Wiley, Hoboken, NJ2005
        • Duval S.
        • Tweedie R.
        Trim and fill: a simple funnel-plot-based method of testing and adjusting for publication bias in meta-analysis.
        Biometrics. 2000; 56: 455-463
        • Kanayama M.
        • Togawa D.
        • Takahashi C.
        • Terai T.
        • Hashimoto T.
        Cross-sectional magnetic resonance imaging study of lumbar disc degeneration in 200 healthy individuals.
        J Neurosurg Spine. 2009; 11: 501-507
        • Södervall J.
        • Karppinen J.
        • Puolitaival J.
        • et al.
        Heart rate variability in sciatica patients referred to spine surgery: a case control study.
        BMC Musculoskelet Disord. 2013; 14: 149
        • Korovessis P.
        • Repantis T.
        • Zacharatos S.
        • Baikousis A.
        Low back pain and sciatica prevalence and intensity reported in a Mediterranean country: ordinal logistic regression analysis.
        Orthopedics. 2012; 35: e1775-e1784
        • Matsudaira K.
        • Kawaguchi M.
        • Isomura T.
        • et al.
        Identification of risk factors for new-onset sciatica in Japanese workers: findings from the Japan epidemiological research of Occupation-related Back pain study.
        Spine (Phila Pa 1976). 2013; 38: E1691-E1700
        • Kääriä S.
        • Leino-Arjas P.
        • Rahkonen O.
        • Lahti J.
        • Lahelma E.
        • Laaksonen M.
        Risk factors of sciatic pain: a prospective study among middle-aged employees.
        Eur J Pain. 2011; 15: 584-590
        • Leino-Arjas P.
        • Solovieva S.
        • Kirjonen J.
        • Reunanen A.
        • Riihimäki H.
        Cardiovascular risk factors and low-back pain in a long-term follow-up of industrial employees.
        Scand J Work Environ Health. 2006; 32: 12-19
        • Miranda H.
        • Viikari-Juntura E.
        • Martikainen R.
        • Takala E.P.
        • Riihimäki H.
        Individual factors, occupational loading, and physical exercise as predictors of sciatic pain.
        Spine (Phila Pa 1976). 2002; 27: 1102-1109
        • Manninen P.
        • Riihimäki H.
        • Heliövaara M.
        Incidence and risk factors of low-back pain in middle-aged farmers.
        Occup Med (Lond). 1995; 45: 141-146
        • Riihimäki H.
        • Viikari-Juntura E.
        • Moneta G.
        • Kuha J.
        • Videman T.
        • Tola S.
        Incidence of sciatic pain among men in machine operating, dynamic physical work, and sedentary work. A three-year follow-up.
        Spine (Phila Pa 1976). 1994; 19: 138-142
        • Riihimäki H.
        • Wickström G.
        • Hänninen K.
        • Luopajärvi T.
        Predictors of sciatic pain among concrete reinforcement workers and house painters—a five-year follow-up.
        Scand J Work Environ Health. 1989; 15: 415-423
        • Karjalainen U.
        • Paananen M.
        • Okuloff A.
        • et al.
        Role of environmental factors and history of low back pain in sciatica symptoms among Finnish adolescents.
        Spine (Phila Pa 1976). 2013; 38: 1105-1111
        • Zhang Y.G.
        • Zhang F.
        • Sun Z.
        • et al.
        A controlled case study of the relationship between environmental risk factors and apoptotic gene polymorphism and lumbar disc herniation.
        Am J Pathol. 2013; 182: 56-63
        • Kostova V.
        • Koleva M.
        Back disorders (low back pain, cervicobrachial and lumbosacral radicular syndromes) and some related risk factors.
        J Neurol Sci. 2001; 192: 17-25
        • Kelsey J.L.
        • Githens P.B.
        • O'Conner T.
        • et al.
        Acute prolapsed lumbar intervertebral disc. An epidemiologic study with special reference to driving automobiles and cigarette smoking.
        Spine (Phila Pa 1976). 1984; 9: 608-613
        • Jhawar B.S.
        • Fuchs C.S.
        • Colditz G.A.
        • Stampfer M.J.
        Cardiovascular risk factors for physician-diagnosed lumbar disc herniation.
        Spine J. 2006; 6: 684-691
        • Violante F.S.
        • Fiori M.
        • Fiorentini C.
        • et al.
        Associations of psychosocial and individual factors with three different categories of back disorder among nursing staff.
        J Occup Health. 2004; 46: 100-108
        • Schumann B.
        • Bolm-Audorff U.
        • Bergmann A.
        • et al.
        Lifestyle factors and lumbar disc disease: results of a German multi-center case-control study (EPILIFT).
        Arthritis Res Ther. 2010; 12: R193
        • Zhang Y.G.
        • Sun Z.
        • Zhang Z.
        • Liu J.
        • Guo X.
        Risk factors for lumbar intervertebral disc herniation in Chinese population: a case-control study.
        Spine (Phila Pa 1976). 2009; 34: E918-E922
        • Heliövaara M.
        • Knekt P.
        • Aromaa A.
        Incidence and risk factors of herniated lumbar intervertebral disc or sciatica leading to hospitalization.
        J Chronic Dis. 1987; 40: 251-258
        • Sørensen I.G.
        • Jacobsen P.
        • Gyntelberg F.
        • Suadicani P.
        Occupational and other predictors of herniated lumbar disc disease-a 33-year follow-up in the Copenhagen male study.
        Spine (Phila Pa 1976). 2011; 36: 1541-1546
        • Wahlström J.
        • Burstrom L.
        • Nilsson T.
        • Järvholm B.
        Risk factors for hospitalization due to lumbar disc disease.
        Spine (Phila Pa 1976). 2012; 37: 1334-1339
        • Kaila-Kangas L.
        • Leino-Arjas P.
        • Riihimäki H.
        • Luukkonen R.
        • Kirjonen J.
        Smoking and overweight as predictors of hospitalization for back disorders.
        Spine (Phila Pa 1976). 2003; 28: 1860-1868
        • Saftic R.
        • Grgic M.
        • Ebling B.
        • Splavski B.
        Case-control study of risk factors for lumbar intervertebral disc herniation in Croatian island populations.
        Croat Med J. 2006; 47: 593-600
        • An H.S.
        • Silveri C.P.
        • Simpson J.M.
        • et al.
        Comparison of smoking habits between patients with surgically confirmed herniated lumbar and cervical disc disease and controls.
        J Spinal Disord. 1994; 7: 369-373
        • Rivinoja A.E.
        • Paananen M.V.
        • Taimela S.P.
        • et al.
        Sports, smoking, and overweight during adolescence as predictors of sciatica in adulthood: a 28-year follow-up study of a birth cohort.
        Am J Epidemiol. 2011; 173: 890-897
        • Mattila V.M.
        • Saarni L.
        • Parkkari J.
        • Koivusilta L.
        • Rimpelä A.
        Early risk factors for lumbar discectomy: an 11-year follow-up of 57,408 adolescents.
        Eur Spine J. 2008; 17: 1317-1323
        • Kauppila L.I.
        • McAlindon T.
        • Evans S.
        • Wilson P.W.
        • Kiel D.
        • Felson D.T.
        Disc degeneration/back pain and calcification of the abdominal aorta. A 25-year follow-up study in Framingham.
        Spine (Phila Pa 1976). 1997; 22 (discussion 1648-9): 1642-1647
        • Shiri R.
        • Viikari-Juntura E.
        • Leino-Arjas P.
        • et al.
        The association between carotid intima-media thickness and sciatica.
        Semin Arthritis Rheum. 2007; 37: 174-181
        • Korkiakoski A.
        • Niinimäki J.
        • Karppinen J.
        • et al.
        Association of lumbar arterial stenosis with low back symptoms: a cross-sectional study using two-dimensional time-of-flight magnetic resonance angiography.
        Acta Radiol. 2009; 50: 48-54
        • Wang D.
        • Nasto L.A.
        • Roughley P.
        • et al.
        Spine degeneration in a murine model of chronic human tobacco smokers.
        Osteoarthritis Cartilage. 2012; 20: 896-905
        • Akmal M.
        • Kesani A.
        • Anand B.
        • Singh A.
        • Wiseman M.
        • Goodship A.
        Effect of nicotine on spinal disc cells: a cellular mechanism for disc degeneration.
        Spine (Phila Pa 1976). 2004; 29: 568-575
        • Uematsu Y.
        • Matuzaki H.
        • Iwahashi M.
        Effects of nicotine on the intervertebral disc: an experimental study in rabbits.
        J Orthop Sci. 2001; 6: 177-182
        • Oda H.
        • Matsuzaki H.
        • Tokuhashi Y.
        • Wakabayashi K.
        • Uematsu Y.
        • Iwahashi M.
        Degeneration of intervertebral discs due to smoking: experimental assessment in a rat-smoking model.
        J Orthop Sci. 2004; 9: 135-141
        • Nemoto Y.
        • Matsuzaki H.
        • Tokuhasi Y.
        • et al.
        Histological changes in intervertebral discs after smoking and cessation: experimental study using a rat passive smoking model.
        J Orthop Sci. 2006; 11: 191-197
        • Haugen A.J.
        • Brox J.I.
        • Grovle L.
        • et al.
        Prognostic factors for non-success in patients with sciatica and disc herniation.
        BMC Musculoskelet Disord. 2012; 13: 183
        • Miwa S.
        • Yokogawa A.
        • Kobayashi T.
        • et al.
        Risk factors of recurrent Lumbar disc herniation: a single center study and review of the literature.
        J Spinal Disord Tech. 2015; 28: E265-E269