INVITED REVIEW
Risk Factors for First Time Incidence Sciatica: A Systematic Review Chad E. Cook1,*, Jeffrey Taylor2, Alexis Wright2, Steven Milosavljevic3, Adam Goode4 & Maureen Whitford1 1
Division of Physical Therapy, Walsh University, North Canton, OH, USA
2
Division of Physical Therapy, High Point University, High Point, NC, USA
3
Division of Physiotherapy, University of Otago, Dunedin, New Zealand
4
Division of Physical Therapy, Duke University, Durham, NC, USA
Abstract Background and Purpose. Characteristically, sciatica involves radiating leg pain that follows a dermatomal pattern along the distribution of the sciatic nerve. To our knowledge, there are no studies that have investigated risk factors associated with first time incidence sciatica. The purpose of the systematic review was to identify the longitudinal risk factors associated with first time incidence sciatica and to report incidence rates for the condition. For the purposes of this review, first time incidence sciatica was defined as either of the following: 1) no prior history of sciatica or 2) transition from a pain-free state to sciatica. Studies included subjects of any age from longitudinal, observational, cohort designs. Methods. The study was a systematic review. Eight of the 239 articles identified by electronic search strategies met the inclusion criteria. Results. Risk factors and their respective effect estimates were reported using descriptive analysis and the preferred reporting items for systematic reviews and meta-analyses guidelines. Modifiable risk factors included smoking, obesity, occupational factors and health status. Non-modifiable factors included age, gender and social class. Incidence rates varied among the included studies, in part reflecting the variability in the operationalized definition of sciatica but ranged from 3 months (OR = 1.8: 1.5–2.2) Men Occupational status Semi-professional workers (OR = 1.5: 1.1–2.1)
Women Occupational status Routine labourer (OR = 1.2: 1.0–1.4) Manual labourer (OR = 1.5: 1.2–1.8) Body mass index Overweight (25.0 = 29.9) (OR = 1.4: 1.2–1.6) Obese (30.0 or greater) (OR = 1.6: 1.3–1.9) Current smoker (OR = 1.5: 1.3–1.8)
Stepwise backwards regression results Past history of LBP (OR = 3.1: 1.6–6.0) >180 cm in height (OR = 2.7: 1.2–6.3) Drive for 2 hours > once a week (OR = 2.7: 1.2–6.4) Bad (score of 5–8) self-rated health (1, excellent to 8, very poor) (OR = 2.8: 1.2–7.1)
Significant predictive risk factors with OR, RR or HR
C. E. Cook et al. Sciatica Incidence
69
70
Miranda et al. (2002)
Manninen et al. (1995)
Study
Table 1. (Continued)
Age: not defined Population: employees of a large forest industry company in Finland
N = 2,077 (53% M and 47% F)
N = 363 (193 M and 170 F) Age: 45–54 years old Population: farmers
Subjects
12 months free at baseline
12 months free from LBP at baseline
Length of pain-free state
1 year
12 years (1 year prevalence)
Follow-up period
Defined as LBP radiating below the knee All 194/2,077 (9.3%)Males: 139/1,387 (10.0%)Females:55/ 493(11.1%)Only reported incidence on 1,880 referents
Defined as LBP within the last 12 months and pain that radiated from either buttock to the foot, so-called sciatic pain All: 35/363 (9.6%) Males: 20/193 (10.3%) Females: 15/170 (8.8%)
Sciatica incidence
(Continues)
Current smoker (OR = 9.6: 1.7–53.0) Adjusted logistic regression with age and gender as confounding variables Age (years) 35–44 (OR = 2.5: 1.4–4.4) 45–54 (OR = 2.4: 1.4–4.2) ≥55 (OR = 3.5: 1.9–6.5) Smoking Ex-smoker (OR = 1.6: 1.1–2.4) Current smoker (1–15 cigarettes/day, 15 years) (OR = 1.9: 1.1–3.5) Current smoker (>15 cigarettes/day, 15 years) (OR = 2.5: 1.6–4.0) Mental stress To some extent (OR = 1.5: 1.0–2.4) Rather much or ‘much’ (OR = 2.6: 1.5–4.5)
Past smoker (OR = 13.1: 2.7–65.1)
Logistic regression adjusted by age, body mass index, body height and mental stress scores
Manual labourer (OR = 2.3: 1.7–3.2) Body mass index Overweight (25.0 = 29.9) (OR = 1.3: 1.0–1.7) Obese (30.0 or greater) (OR = 1.7: 1.1–2.4) Current smoker (OR = 1.5: 1.1–2.1) History of LBP LBP duration ≤3 months (OR = 1.5: 1.2–1.9) LBP duration >3 months (OR = 2.1: 1.2–3.8)
Significant predictive risk factors with OR, RR or HR
Sciatica Incidence C. E. Cook et al.
Physiother. Res. Int. 19 (2014) 65–78 © 2013 John Wiley & Sons, Ltd.
Physiother. Res. Int. 19 (2014) 65–78 © 2013 John Wiley & Sons, Ltd.
Age: 25–49 years old Population: two groups of blue collar workers and one group of white collar workers
N = 1,149 (all men)
Age: 25–49 years old Population: two groups of blue collar workers and one group of white collar workers
N = 1,149 (all men)
Never had sciatic pain
Never had sciatic pain
Same dataset used by Riihimäki et al. (1994).
a
Sciatica defined as low back pain with concurrent leg symptoms.
LBP = low back pain; M = male; F = female; OR = odds ratio; RR = relative risk; HR = hazard ratio.
Pietri-Taleb et al. (1995)†a
Riihimäki et al. (1994)
36 months
36 months
Defined as LBP radiating to the leg (total incidence rate = 19%)
Defined as LBP radiating to the leg incidence rate for machine operators = 22%; incidence rate for carpenters = 24%; incidence rate for office workers = 14%
Regression analysis for while collar workers (N = 426) No predictors Regression analysis for blue collar workers (N = 723) Middlesex Hospital Questionnaire Hysteria Sub-scale-linear term (OR = 1.4: 1.1–1.6)
Occupation Carpenter = (RR = 1.5: 1.1–2.1) Physical exercise More than once a week (RR = 1.3: 1.0–1.6) History of LBP Mild = (RR = 2.7: 1.7–4.2) Severe = (RR = 4.5: 2.7–7.6)
Regression analysis included all occupations (carpenters, machine operators and office workers)
Physical exercise Moderate walking (OR = 1.8: 1.2–2.7) Active walking (OR = 2.2: 1.5–3.4) Work-related factors Moderate twisting of the trunk (OR = 1.7: 1.2–2.4) Much twisting of the trunk (OR = 2.6: 1.7–4.1) Working in kneeling or squatting position >1 hour per day (OR = 2.6: 1.6–4.2) Working with a hand above shoulder level >1 hour per day (OR = 1.9: 1.2–2.9) Working with the trunk forward flexed >2 hours per day (OR = 2.1: 1.4–3.2)
C. E. Cook et al. Sciatica Incidence
71
72
Population: 1,966 northern Finland birth cohort
Age: 41 years old when re-assessed
N = 9,016 14 year old adolescents
Age: 15 years old and older Population: cohort of Finnish adults, initially examined by the mobile clinic of the Social Insurance Institution
N = 57,000 (baseline gender data not reported)
Subjects
Sciatica defined by hospitalization for the condition.
OR = odds ratio; RR = relative risk; HR = hazard ratio.
Rivinoja et al. (2011)
Heliövaara et al. (1987)
Study
Table 2. Risk factors for sciatica in longitudinal studies
Responded as no sciatica at age 14 years
At baseline date only
Length of pain-free state
27 years
11 years
Follow-up period
Incidence rate: all = 4.1%; females = 119/373 and males = 254/373
Defined as surgery associated with sciatica. Also looked at combined first time LBP and sciatica and second hospitalization with sciatica
Of the 371 reported: males: 226/371 Females: 145/371
All: 371/57,000 (0.65%)
Defined as hospitalization
Sciatica Incidence
Male gender (RR = 1.3) Age at entry (years) 20–29 (RR = 2.7) 30–39 (RR = 8.6) 40–49 (RR = 12.1) 50–59 (RR = 7.3) 60 and older (RR = 4.3) Type of population Industrial (RR = 1.5) Occupational group Agriculture/forestry (RR = 1.5) Industry (RR = 1.9) Service (RR = 2.2) Social class (income) II (second highest) (RR = 3.7) III (middle class) (RR = 4.0) IV (second lowest) (RR = 3.1) V (lowest)(RR = 1.9) Chronic cough (RR = 1.5) Number of distress symptoms One total (RR = 1.4) Regression analysis for first hospitalization and surgery for sciatica Women Nothing Men Nothing Regression analysis for second hospitalization and surgery for sciatica Women Overweight/obesity = (HR = 7.1: 1.5–34.4) Males Smoking = (HR = 3.2: 1.2–8.2)
Significant predictive risk factors with OR, RR or HR
Sciatica Incidence C. E. Cook et al.
Physiother. Res. Int. 19 (2014) 65–78 © 2013 John Wiley & Sons, Ltd.
Physiother. Res. Int. 19 (2014) 65–78 © 2013 John Wiley & Sons, Ltd.
Low Low High Low Low Low Low Low
High
High
High
High
High
High
Low
Sampling frame was a close representation of population
High
Low
Low
Low
Low
Low
Low
Low
Low nonresponse bias
High
Low
Low
Low
High
Low
Low
Low
Data collected directly from the subjects
High
High
High
High
High
High
High
High
Acceptable case definition was used
High
Low
Low
Low
High
Low
Low
High
Instrumentation used was reliable and valid
High risk of bias: Further research is very likely to have an important impact on our confidence in the estimate and is likely to change the estimate.
Low
Low
Low
Low
Low
Low
Low
Low
The same mode of data collection was used for all studies
Internal validity
Moderate risk of bias: Further research is likely to have an important impact on our confidence in the estimate and may change the estimate.
Low risk of bias: Further research is very unlikely to change our confidence in the estimate.
Low
High
High
High
High
High
High
Low
Randomized sample or census data were used
External validity
Low
Population was a close representation of national population
Yes, low risk; no, high risk.
Heliövaara et al. (1987) Kääriä et al. (2011) LeClerc et al. (2003) Manninen et al. (1995) Miranda et al. (2002) Riihimäki et al. (1994) Pietri-Taleb et al. (1995) Rivinoja et al. (2011)
Article
Table 3. Risk of bias scores for included studies
Low
Low
Low
High
Low
High
Low
Low
The length of the assessment period was appropriate
Low
High
High
Low
High
Low
Low
Low
The numerator and denominator of the parameter of interest was reported Low risk of bias Low risk of bias Moderate risk of bias High risk of bias Moderate risk of bias Moderate risk of bias Moderate risk of bias Moderate risk of bias
Overall score
C. E. Cook et al. Sciatica Incidence
73
C. E. Cook et al.
Sciatica Incidence
in which a pain-free population was followed longitudinally to first time incidence sciatica. Our intention was to identify consistent risk factors, especially modifiable risk factors that could be targeted for primary prevention. Age and history of LBP were the most common non-modifiable risk factors found in the studies. Although the same baseline risk factors were not examined in all studies, the majority investigated modifiable factors such as work related activities, smoking and obesity/overweight and all three of these factors were typically associated with higher rates of sciatica. Within this discussion, we will outline the consistency of our risk factor findings with the current literature and will discuss these findings in the context of the precincts of the search strategy and accepted manuscripts.
or a collective risk factor with other conditions such as poor health, obesity and LBP. Past studies have shown some success through the use of smoking cessation programmes. Smoking cessation has benefited from pharmacologic developments, increased public awareness of the health risks affiliated with smoking, negative advertising for smoking, changes in public opinions about acceptance of second-hand smoking (Black, 2010) and sophistication of formal stop-smoking programmes (Batra, 2011). Results have shown up to 40% success rates for selected populations (Batra, 2011), although non-compliance to abstinence of smoking is especially high even among the most ill (Casillas et al. 2011). To our knowledge, there are no studies that have directly measured the effectiveness of physiotherapy led smoking sensation programmes.
Risk factor: smoking status
Risk factor: obesity/overweight
Smoking status, which is a modifiable life choice, was identified by four of eight studies as a risk factor for sciatica, and interestingly, being a smoker and being an ex-smoker were both identified as risk factors for sciatica. Smoking has also been identified as a risk factor in a cross-sectional study associated with disc-related sciatica (Younes et al., 2006), although a recent review demonstrated that smoking was not associated with prognosis during conservative care for sciatica (Ashworth et al., 2011). Smoking has been associated with chronic pain syndromes, including sciatica, in a recent investigation of over 6,000 women (Mitchell et al., 2011). The deleterious effects of smoking can last anywhere from 2 to 30 years following cessation, which may explain why both smoking and a history of smoking (i.e. ex-smokers) are listed as risk factors for sciatica (Qiao et al., 2000). However, an exact connection between smoking and sciatica is less absolute. Increased propensity to cough is often associated with smoking and was found as a risk factor in one study (Heliövaara et al., 1987). Additionally, decreased pain threshold through sensitizing pain receptors from nicotine is hypothesized to lead to the possible relationship between sciatica and smoking (Mitchell et al., 2011). Another hypothesis suggests that hyperlipidemia, a side effect of smoking and poor health, may contribute to atherosclerosis sciatica, which is understood to be is associated with leg pain (Leino-Arjas et al., 2008); however, further research is needed to determine if smoking is directly associated with this
Obesity/overweight status was an identified risk factor in two of the reviewed studies. Obesity has been associated with juvenile disc degeneration (14-fold greater prevalence), and individuals with disc degeneration have been shown to have an increased prevalence of sciatica (Samartzis et al., 2011). van Nieuwenhuyse et al. (2009) found a twofold increased risk of developing LBP in obese individuals as compared with their non-obese colleagues. In a relatively recent review, obesity was associated with clinically defined sciatica in most cohort and case control studies (Shiri et al., 2007). The authors of that review did not find a direct dose–response relationship related to weight but rather proposed that being overweight may increase systemic inflammatory responses overall (increased C-reactive proteins) (Shiri et al., 2007), leading to clinically diagnosed sciatica. This is consistent with previous reviews that reiterated that obesity is not necessarily causally associated with sciatica (Heliovaara, 1989). In addition to its general association with sciatica, obesity has been identified as an associative factor for poor outcomes in patients with mechanical sciatica (Bejia et al., 2004), further emphasizing its importance in preventative management. Obesity is also a modifiable risk factor, and lifestyle modification programmes have been shown to be viable for primary care (Ma et al., 2013). Modifying obesity requires behavioural changes, increases in physical activity and appropriate diet regulations. Although physiotherapists have been recommended for primary prevention of obesity, to our knowledge there are no trials that have
74
Physiother. Res. Int. 19 (2014) 65–78 © 2013 John Wiley & Sons, Ltd.
C. E. Cook et al.
examined the effectiveness of dedicated weight loss interventions. At present, preventative elements throughout the world are underrated and undervalued by public health agencies (Wittink et al., 2011).
Risk factor: work-related activities Work-related activity within manual occupational classifications has considerable evidence for being associated with occupational LBP. The association, however, is usually more strongly evident in multifactorial models. There is a wealth of literature and arguments within differing research disciplines that encompass most of the bio-psychosocial spectrum, and thus, it is often difficult to tease out the predominant biomechanical, psychological or social factor linked to the development of LBP (Marras, 2005). Development of occupational LBP has been linked with many factors including various personal factors, anthropometry, various psychological factors, loading, work postures and cumulative exposures. Although sciatica may or may not be associated with LBP, as previously stated, it is thought that 90% of cases of sciatica is caused by a lumbar herniated disc with subsequent nerve root compression (Koes et al., 2007). Recent research identifying possible links between personality profiles and immunological responses lends support to the premise that the condition of sciatica is complex. Furthermore, it supports the need to develop well-powered multifactorial cohort studies to identify the factors most likely linked to development of occupational LBP (Splittstoesser et al., 2012). The strong associations of sciatica with a history of manual labour and previous episodes of LBP imply that high number of exposures to physical loading are likely to set up the disc (or the spinal segment) for some form of physical insult (e.g. herniation). This places the normal physical interface between neural and articular tissues in a pathological state with consequential neural irritation and subsequent sciatic pain. Evidence for physical exposures relative to sciatica is currently weak; however, awkward twisted work postures, exposure to whole body vibration, load lifting and multiaxial cumulative loading exposures are consistently associated with occupational LBP (Miranda et al., 2002). It may be that over time, excessive physical exposures within the constraints of other bio-psychosocial factors (Leclerc et al., 2003) are a triggering factor in those who develop sciatica within the specific occupations. Physiother. Res. Int. 19 (2014) 65–78 © 2013 John Wiley & Sons, Ltd.
Sciatica Incidence
Interestingly, recent evidence argues that physical exposures are probably linked to the development of the sciatica complaint, but it is the psychosocial factors that are most strongly associated with its persistence over time (Miranda et al., 2002). Modification of the workplace environment for LBP/sciatica has primarily addressed the following: 1) back belts; 2) education and task modification; and 3) education and task modification with workstation redesign. Positive outcomes have been associated with modifications that focused on compliance and occupational training (Gatty et al., 2003). However, to date, the benefits associated with primary prevention in this genre and others remains largely undetermined, because risk factor modification will not necessarily achieve a complete 100% prevention. In fact, overall, there is limited robust evidence for numerous aspects of prevention in LBP (Burton et al., 2006). Possible explanations of our findings As with most longitudinal studies that measure the influence of baseline variables toward a dedicated outcome, the findings are only as pertinent as the consistency of the study methodology. There was noticeable variability in the operational definitions of the diagnosis of sciatica used within the eight accepted papers; a finding that may reflect the lack of consistent diagnostic criteria within the literature (Konstantinou and Dunn, 2008; Lewis et al., 2011). Within six of eight studies, the description of sciatica involved LBP radiating below the knee or to a definitive or non-definitive area within the lower extremity. In two other studies, sciatica was defined as a medical diagnosis used during hospital admission. The inconsistent operational definitions are a likely reason for the wide range of reported incidence rates among the eight studies. Overall, incidence rates ranged from
Risk Factors for First Time Incidence Sciatica: A Systematic Review Chad E. Cook1,*, Jeffrey Taylor2, Alexis Wright2, Steven Milosavljevic3, Adam Goode4 & Maureen Whitford1 1
Division of Physical Therapy, Walsh University, North Canton, OH, USA
2
Division of Physical Therapy, High Point University, High Point, NC, USA
3
Division of Physiotherapy, University of Otago, Dunedin, New Zealand
4
Division of Physical Therapy, Duke University, Durham, NC, USA
Abstract Background and Purpose. Characteristically, sciatica involves radiating leg pain that follows a dermatomal pattern along the distribution of the sciatic nerve. To our knowledge, there are no studies that have investigated risk factors associated with first time incidence sciatica. The purpose of the systematic review was to identify the longitudinal risk factors associated with first time incidence sciatica and to report incidence rates for the condition. For the purposes of this review, first time incidence sciatica was defined as either of the following: 1) no prior history of sciatica or 2) transition from a pain-free state to sciatica. Studies included subjects of any age from longitudinal, observational, cohort designs. Methods. The study was a systematic review. Eight of the 239 articles identified by electronic search strategies met the inclusion criteria. Results. Risk factors and their respective effect estimates were reported using descriptive analysis and the preferred reporting items for systematic reviews and meta-analyses guidelines. Modifiable risk factors included smoking, obesity, occupational factors and health status. Non-modifiable factors included age, gender and social class. Incidence rates varied among the included studies, in part reflecting the variability in the operationalized definition of sciatica but ranged from 3 months (OR = 1.8: 1.5–2.2) Men Occupational status Semi-professional workers (OR = 1.5: 1.1–2.1)
Women Occupational status Routine labourer (OR = 1.2: 1.0–1.4) Manual labourer (OR = 1.5: 1.2–1.8) Body mass index Overweight (25.0 = 29.9) (OR = 1.4: 1.2–1.6) Obese (30.0 or greater) (OR = 1.6: 1.3–1.9) Current smoker (OR = 1.5: 1.3–1.8)
Stepwise backwards regression results Past history of LBP (OR = 3.1: 1.6–6.0) >180 cm in height (OR = 2.7: 1.2–6.3) Drive for 2 hours > once a week (OR = 2.7: 1.2–6.4) Bad (score of 5–8) self-rated health (1, excellent to 8, very poor) (OR = 2.8: 1.2–7.1)
Significant predictive risk factors with OR, RR or HR
C. E. Cook et al. Sciatica Incidence
69
70
Miranda et al. (2002)
Manninen et al. (1995)
Study
Table 1. (Continued)
Age: not defined Population: employees of a large forest industry company in Finland
N = 2,077 (53% M and 47% F)
N = 363 (193 M and 170 F) Age: 45–54 years old Population: farmers
Subjects
12 months free at baseline
12 months free from LBP at baseline
Length of pain-free state
1 year
12 years (1 year prevalence)
Follow-up period
Defined as LBP radiating below the knee All 194/2,077 (9.3%)Males: 139/1,387 (10.0%)Females:55/ 493(11.1%)Only reported incidence on 1,880 referents
Defined as LBP within the last 12 months and pain that radiated from either buttock to the foot, so-called sciatic pain All: 35/363 (9.6%) Males: 20/193 (10.3%) Females: 15/170 (8.8%)
Sciatica incidence
(Continues)
Current smoker (OR = 9.6: 1.7–53.0) Adjusted logistic regression with age and gender as confounding variables Age (years) 35–44 (OR = 2.5: 1.4–4.4) 45–54 (OR = 2.4: 1.4–4.2) ≥55 (OR = 3.5: 1.9–6.5) Smoking Ex-smoker (OR = 1.6: 1.1–2.4) Current smoker (1–15 cigarettes/day, 15 years) (OR = 1.9: 1.1–3.5) Current smoker (>15 cigarettes/day, 15 years) (OR = 2.5: 1.6–4.0) Mental stress To some extent (OR = 1.5: 1.0–2.4) Rather much or ‘much’ (OR = 2.6: 1.5–4.5)
Past smoker (OR = 13.1: 2.7–65.1)
Logistic regression adjusted by age, body mass index, body height and mental stress scores
Manual labourer (OR = 2.3: 1.7–3.2) Body mass index Overweight (25.0 = 29.9) (OR = 1.3: 1.0–1.7) Obese (30.0 or greater) (OR = 1.7: 1.1–2.4) Current smoker (OR = 1.5: 1.1–2.1) History of LBP LBP duration ≤3 months (OR = 1.5: 1.2–1.9) LBP duration >3 months (OR = 2.1: 1.2–3.8)
Significant predictive risk factors with OR, RR or HR
Sciatica Incidence C. E. Cook et al.
Physiother. Res. Int. 19 (2014) 65–78 © 2013 John Wiley & Sons, Ltd.
Physiother. Res. Int. 19 (2014) 65–78 © 2013 John Wiley & Sons, Ltd.
Age: 25–49 years old Population: two groups of blue collar workers and one group of white collar workers
N = 1,149 (all men)
Age: 25–49 years old Population: two groups of blue collar workers and one group of white collar workers
N = 1,149 (all men)
Never had sciatic pain
Never had sciatic pain
Same dataset used by Riihimäki et al. (1994).
a
Sciatica defined as low back pain with concurrent leg symptoms.
LBP = low back pain; M = male; F = female; OR = odds ratio; RR = relative risk; HR = hazard ratio.
Pietri-Taleb et al. (1995)†a
Riihimäki et al. (1994)
36 months
36 months
Defined as LBP radiating to the leg (total incidence rate = 19%)
Defined as LBP radiating to the leg incidence rate for machine operators = 22%; incidence rate for carpenters = 24%; incidence rate for office workers = 14%
Regression analysis for while collar workers (N = 426) No predictors Regression analysis for blue collar workers (N = 723) Middlesex Hospital Questionnaire Hysteria Sub-scale-linear term (OR = 1.4: 1.1–1.6)
Occupation Carpenter = (RR = 1.5: 1.1–2.1) Physical exercise More than once a week (RR = 1.3: 1.0–1.6) History of LBP Mild = (RR = 2.7: 1.7–4.2) Severe = (RR = 4.5: 2.7–7.6)
Regression analysis included all occupations (carpenters, machine operators and office workers)
Physical exercise Moderate walking (OR = 1.8: 1.2–2.7) Active walking (OR = 2.2: 1.5–3.4) Work-related factors Moderate twisting of the trunk (OR = 1.7: 1.2–2.4) Much twisting of the trunk (OR = 2.6: 1.7–4.1) Working in kneeling or squatting position >1 hour per day (OR = 2.6: 1.6–4.2) Working with a hand above shoulder level >1 hour per day (OR = 1.9: 1.2–2.9) Working with the trunk forward flexed >2 hours per day (OR = 2.1: 1.4–3.2)
C. E. Cook et al. Sciatica Incidence
71
72
Population: 1,966 northern Finland birth cohort
Age: 41 years old when re-assessed
N = 9,016 14 year old adolescents
Age: 15 years old and older Population: cohort of Finnish adults, initially examined by the mobile clinic of the Social Insurance Institution
N = 57,000 (baseline gender data not reported)
Subjects
Sciatica defined by hospitalization for the condition.
OR = odds ratio; RR = relative risk; HR = hazard ratio.
Rivinoja et al. (2011)
Heliövaara et al. (1987)
Study
Table 2. Risk factors for sciatica in longitudinal studies
Responded as no sciatica at age 14 years
At baseline date only
Length of pain-free state
27 years
11 years
Follow-up period
Incidence rate: all = 4.1%; females = 119/373 and males = 254/373
Defined as surgery associated with sciatica. Also looked at combined first time LBP and sciatica and second hospitalization with sciatica
Of the 371 reported: males: 226/371 Females: 145/371
All: 371/57,000 (0.65%)
Defined as hospitalization
Sciatica Incidence
Male gender (RR = 1.3) Age at entry (years) 20–29 (RR = 2.7) 30–39 (RR = 8.6) 40–49 (RR = 12.1) 50–59 (RR = 7.3) 60 and older (RR = 4.3) Type of population Industrial (RR = 1.5) Occupational group Agriculture/forestry (RR = 1.5) Industry (RR = 1.9) Service (RR = 2.2) Social class (income) II (second highest) (RR = 3.7) III (middle class) (RR = 4.0) IV (second lowest) (RR = 3.1) V (lowest)(RR = 1.9) Chronic cough (RR = 1.5) Number of distress symptoms One total (RR = 1.4) Regression analysis for first hospitalization and surgery for sciatica Women Nothing Men Nothing Regression analysis for second hospitalization and surgery for sciatica Women Overweight/obesity = (HR = 7.1: 1.5–34.4) Males Smoking = (HR = 3.2: 1.2–8.2)
Significant predictive risk factors with OR, RR or HR
Sciatica Incidence C. E. Cook et al.
Physiother. Res. Int. 19 (2014) 65–78 © 2013 John Wiley & Sons, Ltd.
Physiother. Res. Int. 19 (2014) 65–78 © 2013 John Wiley & Sons, Ltd.
Low Low High Low Low Low Low Low
High
High
High
High
High
High
Low
Sampling frame was a close representation of population
High
Low
Low
Low
Low
Low
Low
Low
Low nonresponse bias
High
Low
Low
Low
High
Low
Low
Low
Data collected directly from the subjects
High
High
High
High
High
High
High
High
Acceptable case definition was used
High
Low
Low
Low
High
Low
Low
High
Instrumentation used was reliable and valid
High risk of bias: Further research is very likely to have an important impact on our confidence in the estimate and is likely to change the estimate.
Low
Low
Low
Low
Low
Low
Low
Low
The same mode of data collection was used for all studies
Internal validity
Moderate risk of bias: Further research is likely to have an important impact on our confidence in the estimate and may change the estimate.
Low risk of bias: Further research is very unlikely to change our confidence in the estimate.
Low
High
High
High
High
High
High
Low
Randomized sample or census data were used
External validity
Low
Population was a close representation of national population
Yes, low risk; no, high risk.
Heliövaara et al. (1987) Kääriä et al. (2011) LeClerc et al. (2003) Manninen et al. (1995) Miranda et al. (2002) Riihimäki et al. (1994) Pietri-Taleb et al. (1995) Rivinoja et al. (2011)
Article
Table 3. Risk of bias scores for included studies
Low
Low
Low
High
Low
High
Low
Low
The length of the assessment period was appropriate
Low
High
High
Low
High
Low
Low
Low
The numerator and denominator of the parameter of interest was reported Low risk of bias Low risk of bias Moderate risk of bias High risk of bias Moderate risk of bias Moderate risk of bias Moderate risk of bias Moderate risk of bias
Overall score
C. E. Cook et al. Sciatica Incidence
73
C. E. Cook et al.
Sciatica Incidence
in which a pain-free population was followed longitudinally to first time incidence sciatica. Our intention was to identify consistent risk factors, especially modifiable risk factors that could be targeted for primary prevention. Age and history of LBP were the most common non-modifiable risk factors found in the studies. Although the same baseline risk factors were not examined in all studies, the majority investigated modifiable factors such as work related activities, smoking and obesity/overweight and all three of these factors were typically associated with higher rates of sciatica. Within this discussion, we will outline the consistency of our risk factor findings with the current literature and will discuss these findings in the context of the precincts of the search strategy and accepted manuscripts.
or a collective risk factor with other conditions such as poor health, obesity and LBP. Past studies have shown some success through the use of smoking cessation programmes. Smoking cessation has benefited from pharmacologic developments, increased public awareness of the health risks affiliated with smoking, negative advertising for smoking, changes in public opinions about acceptance of second-hand smoking (Black, 2010) and sophistication of formal stop-smoking programmes (Batra, 2011). Results have shown up to 40% success rates for selected populations (Batra, 2011), although non-compliance to abstinence of smoking is especially high even among the most ill (Casillas et al. 2011). To our knowledge, there are no studies that have directly measured the effectiveness of physiotherapy led smoking sensation programmes.
Risk factor: smoking status
Risk factor: obesity/overweight
Smoking status, which is a modifiable life choice, was identified by four of eight studies as a risk factor for sciatica, and interestingly, being a smoker and being an ex-smoker were both identified as risk factors for sciatica. Smoking has also been identified as a risk factor in a cross-sectional study associated with disc-related sciatica (Younes et al., 2006), although a recent review demonstrated that smoking was not associated with prognosis during conservative care for sciatica (Ashworth et al., 2011). Smoking has been associated with chronic pain syndromes, including sciatica, in a recent investigation of over 6,000 women (Mitchell et al., 2011). The deleterious effects of smoking can last anywhere from 2 to 30 years following cessation, which may explain why both smoking and a history of smoking (i.e. ex-smokers) are listed as risk factors for sciatica (Qiao et al., 2000). However, an exact connection between smoking and sciatica is less absolute. Increased propensity to cough is often associated with smoking and was found as a risk factor in one study (Heliövaara et al., 1987). Additionally, decreased pain threshold through sensitizing pain receptors from nicotine is hypothesized to lead to the possible relationship between sciatica and smoking (Mitchell et al., 2011). Another hypothesis suggests that hyperlipidemia, a side effect of smoking and poor health, may contribute to atherosclerosis sciatica, which is understood to be is associated with leg pain (Leino-Arjas et al., 2008); however, further research is needed to determine if smoking is directly associated with this
Obesity/overweight status was an identified risk factor in two of the reviewed studies. Obesity has been associated with juvenile disc degeneration (14-fold greater prevalence), and individuals with disc degeneration have been shown to have an increased prevalence of sciatica (Samartzis et al., 2011). van Nieuwenhuyse et al. (2009) found a twofold increased risk of developing LBP in obese individuals as compared with their non-obese colleagues. In a relatively recent review, obesity was associated with clinically defined sciatica in most cohort and case control studies (Shiri et al., 2007). The authors of that review did not find a direct dose–response relationship related to weight but rather proposed that being overweight may increase systemic inflammatory responses overall (increased C-reactive proteins) (Shiri et al., 2007), leading to clinically diagnosed sciatica. This is consistent with previous reviews that reiterated that obesity is not necessarily causally associated with sciatica (Heliovaara, 1989). In addition to its general association with sciatica, obesity has been identified as an associative factor for poor outcomes in patients with mechanical sciatica (Bejia et al., 2004), further emphasizing its importance in preventative management. Obesity is also a modifiable risk factor, and lifestyle modification programmes have been shown to be viable for primary care (Ma et al., 2013). Modifying obesity requires behavioural changes, increases in physical activity and appropriate diet regulations. Although physiotherapists have been recommended for primary prevention of obesity, to our knowledge there are no trials that have
74
Physiother. Res. Int. 19 (2014) 65–78 © 2013 John Wiley & Sons, Ltd.
C. E. Cook et al.
examined the effectiveness of dedicated weight loss interventions. At present, preventative elements throughout the world are underrated and undervalued by public health agencies (Wittink et al., 2011).
Risk factor: work-related activities Work-related activity within manual occupational classifications has considerable evidence for being associated with occupational LBP. The association, however, is usually more strongly evident in multifactorial models. There is a wealth of literature and arguments within differing research disciplines that encompass most of the bio-psychosocial spectrum, and thus, it is often difficult to tease out the predominant biomechanical, psychological or social factor linked to the development of LBP (Marras, 2005). Development of occupational LBP has been linked with many factors including various personal factors, anthropometry, various psychological factors, loading, work postures and cumulative exposures. Although sciatica may or may not be associated with LBP, as previously stated, it is thought that 90% of cases of sciatica is caused by a lumbar herniated disc with subsequent nerve root compression (Koes et al., 2007). Recent research identifying possible links between personality profiles and immunological responses lends support to the premise that the condition of sciatica is complex. Furthermore, it supports the need to develop well-powered multifactorial cohort studies to identify the factors most likely linked to development of occupational LBP (Splittstoesser et al., 2012). The strong associations of sciatica with a history of manual labour and previous episodes of LBP imply that high number of exposures to physical loading are likely to set up the disc (or the spinal segment) for some form of physical insult (e.g. herniation). This places the normal physical interface between neural and articular tissues in a pathological state with consequential neural irritation and subsequent sciatic pain. Evidence for physical exposures relative to sciatica is currently weak; however, awkward twisted work postures, exposure to whole body vibration, load lifting and multiaxial cumulative loading exposures are consistently associated with occupational LBP (Miranda et al., 2002). It may be that over time, excessive physical exposures within the constraints of other bio-psychosocial factors (Leclerc et al., 2003) are a triggering factor in those who develop sciatica within the specific occupations. Physiother. Res. Int. 19 (2014) 65–78 © 2013 John Wiley & Sons, Ltd.
Sciatica Incidence
Interestingly, recent evidence argues that physical exposures are probably linked to the development of the sciatica complaint, but it is the psychosocial factors that are most strongly associated with its persistence over time (Miranda et al., 2002). Modification of the workplace environment for LBP/sciatica has primarily addressed the following: 1) back belts; 2) education and task modification; and 3) education and task modification with workstation redesign. Positive outcomes have been associated with modifications that focused on compliance and occupational training (Gatty et al., 2003). However, to date, the benefits associated with primary prevention in this genre and others remains largely undetermined, because risk factor modification will not necessarily achieve a complete 100% prevention. In fact, overall, there is limited robust evidence for numerous aspects of prevention in LBP (Burton et al., 2006). Possible explanations of our findings As with most longitudinal studies that measure the influence of baseline variables toward a dedicated outcome, the findings are only as pertinent as the consistency of the study methodology. There was noticeable variability in the operational definitions of the diagnosis of sciatica used within the eight accepted papers; a finding that may reflect the lack of consistent diagnostic criteria within the literature (Konstantinou and Dunn, 2008; Lewis et al., 2011). Within six of eight studies, the description of sciatica involved LBP radiating below the knee or to a definitive or non-definitive area within the lower extremity. In two other studies, sciatica was defined as a medical diagnosis used during hospital admission. The inconsistent operational definitions are a likely reason for the wide range of reported incidence rates among the eight studies. Overall, incidence rates ranged from
