Core Stability Exercises for Low Back Pain in Athletes

A Systematic Review of the Literature

Abstract

Objective: 

The aim of this study was to systematically review the evidence for the effectiveness of core stability exercises for treating athletes with low back pain (LBP).

Data Sources: 

We searched several databases (Medline, AMED, CINAHL, SportDiscus, and EMBASE). Our eligibility criteria consisted of articles published in a peer-reviewed journal in English, using any prospective clinical study design, where athletes with nonspecific LBP were treated with core stability exercises in at least 1 study arm, and back pain intensity and/or disability were used as outcome measures. All included randomized controlled trials (RCTs) were assessed for risk of bias using the Cochrane Risk of Bias tool, whereas non-RCT studies were assessed for quality using the Downs and Black checklist.

Main Results: 

Five studies including 151 participants met the inclusion criteria, including 2 RCTs. The quality of the literature on this topic was deemed to be low overall, with only 1 non-RCT having a moderate quality score, and 1 RCT having a lower risk of bias. Four studies reported statistically significant decreases in back pain intensity in their core stability intervention group.

Conclusions: 

The quantity and quality of literature on the use of core stability exercises for treating LBP in athletes is low. The existing evidence has been conducted on small and heterogeneous study populations using interventions that vary drastically with only mixed results and short-term follow-up. This precludes the formulation of strong conclusions, and additional high quality research is clearly needed.

INTRODUCTION

Low back pain (LBP) is common in athletes. The 1-year prevalence of LBP has been reported to be 68% among top athletes from multiple sports;¹ studies on specific sports have reported 1-year prevalence of 49% in orienteering,² 54% in wrestlers,¹ 86% in gymnasts,³ and 94% in hockey players.¹ Comparatively, the reported 1-year prevalence of LBP in the general population varies from 22% to 65%.⁴ Although the results of 2 studies suggest that the prevalence of LBP in athletes is similar to that of the general population,^2,5 1 study investigating the prevalence of LBP among Australian Rules football players indicated that it was higher than in a nonathletic control group.⁶

Depending on the sport, athletes require their lower backs to have the capacity to tolerate high loads and perform complex repetitive movements.^7–9 It is well documented that certain forms of spinal injury are more common in athletes than in the general population. For example, 33% of elite American football linemen were found to have hyperconcavity of the vertebral endplates with expansion of the disk space compared with 8% of age-matched controls,¹⁰ although this does not seem to have an effect on lumbosacral spine symptoms or duration of playing career.¹¹ Degenerative disc disease is more common in elite athletes than nonathletes.^12–14 Spondylolytic injuries seem to be more common in some sports than others (eg, American football, rowing, diving, gymnastics, martial arts, rugby), with throwing sports, in particular, having higher rates of spondylolytic injuries.^7,14–18 American football linemen produce nearly 7 times their body weight in compressive force and 2.6 times their body weight in shear force on the L4–L5 segment when blocking; with repetition, these forces may increase the risk for fatigue failure and injury.⁷ Cricket bowlers place high compressive loads on their spines during the power phase of the throw because of the frequent repetition of high velocity lumbar movements in a very specific sequence.^9,18 Gymnasts also place large vertical and lateral impact forces on their spines,¹⁹ and overuse has been implicated as a potential risk factor for back injury in this population.¹⁸

There are many possible treatment options for LBP in athletes including medications, biopsychosocial interventions, physical and electrical modalities, manual therapies, and exercise therapies.²⁰ Among exercise therapies, there are different forms of exercise that can be prescribed for LBP, including stretching or mobility exercises, cardiovascular endurance or aerobic exercises, and strengthening exercises.²¹ One form of strengthening exercise that has received increasing attention in the literature and popular media is “core stability exercise.” This type of exercise is variably defined as any exercises that strengthen spinal musculature or specifically as those that emphasize the deep lumbopelvic musculature (eg, transversus abdominis, multifidus).²² There is some evidence that exercises thought to target these deep muscles are effective in treating chronic LBP in the general population.^23,24 However, it has also been argued that focusing on 1 or 2 deep muscles in core stability exercise programs is misguided because both deep and superficial muscles contribute to spinal stability.^25–29 McGill et al²⁸ state that “any exercise that grooves motor patterns, that ensure a stable spine, through repetition, constitutes a ‘stabilization’ exercise.” A recent meta-analysis that defined core stability exercise broadly as “the reinforcement of the ability to insure stability of the neutral spine position” reported that such exercises are more effective than general exercise in treating chronic LBP in the general population.³⁰

Differences in the athlete compared with the general population (eg, higher physical demands, repetitive motions, and higher baseline fitness level) warrant assessment of treatment evidence specifically for this population. The aim of this study was to systematically review the available evidence for the effectiveness of core stability exercises for athletes with LBP.

METHODS

We conducted and reported this systematic review according to the PRISMA guidelines,³¹ with a protocol defined a priori. We searched 5 online databases (Medline, AMED, CINAHL, SportDiscus, and EMBASE) from the start date of the respective database through the last week of June 2012 with no language restrictions, using a comprehensive set of MeSH and key word terms referring to athletes, LBP, and core stability exercises (Appendix for the Medline search strategy). Two reviewers (K.J.S. and S.S.) independently screened the titles and abstracts for potentially applicable studies, and full-text reports of potentially eligible citations were retrieved for final selection. Any disagreement on inclusion was discussed and mediated by a third author (J.A.H.) when necessary. We searched reference lists of all included studies to identify other potentially relevant studies. Included studies had to meet specific criteria (Table 1). Two of the authors (P.B. and S.S.) independently extracted data from the included studies into a prespecified electronic data extraction form, which was subsequently reviewed by a third author (J.A.H.).

TABLE 1: Study Criteria for Inclusion in the Review

Critical appraisal of included studies was conducted independently by 2 authors (K.J.S. and P.B.) with a third author (J.A.H.) resolving any differences. All included randomized controlled trials (RCTs) were evaluated using the Cochrane Back Review Group modification of the Cochrane risk of bias tool.³² Randomized controlled trials were deemed to have a low risk of bias if they had ≥6 of the 12 risk of bias criteria scored as low risk.³³ Non-RCT studies were assessed for quality using the Downs and Black checklist.³⁴ Non-RCT studies were deemed to be high quality for scores of ≥21 on the checklist, moderate quality for scores between 11 and 20, and low quality if they scored ≤10.³⁵ Statistical significance (set at P < 0.05) for between-group differences and within-group differences over time was used to interpret positive results. To determine whether minimal clinically important differences were achieved between groups in the included studies, we considered a 20-point (out of 100) or 2-point (out of 10) improvement in pain intensity and a 10-point (out of 100) improvement in functional measures to be clinically important.³⁶ We also considered a minimum average of 30% improvement within groups from baseline to be clinically important.³⁶

Two of the authors (K.J.S. and P.B.) independently applied the GRADE criteria³⁷ for interpreting the overall quality of the available evidence. Using this framework, high quality evidence indicates that further research is very unlikely to change our confidence in the estimate of effect, moderate quality evidence indicates that further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate, low quality evidence indicates that further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate, and very low quality evidence indicates that any estimate of effect is very uncertain.

RESULTS

Search Results and Study Designs

We identified 677 potential citations and screened 9 full-text articles (Figure). Four of the 9 articles were excluded because of poor reporting of the LBP subgroup and a lack of pain intensity reporting³⁸ or because the intervention did not specifically include core stability exercises.^38–41 Five studies met all selection criteria and were included in the review: 3 observational studies^42–44 and 2 RCTs.^45,46Table 2 presents a description of included studies.

FIGURE: Selection process of included articles.

TABLE 2: Summary of Included Studies

Participants

Sample sizes of included studies ranged from 7⁴⁴ to 50 (25 participants per group).⁴² All of the participants in 1 study were women,⁴³ whereas only men participated in the remaining studies. In 2 studies, the participants were middle-aged and older.^42,46 The participants were adolescents in 1 study,⁴³ whereas in the remaining 2 studies, the average age of the patients was in their early twenties.^44,45 Sports involved included cricket,⁴⁴ field hockey,⁴⁵ ice hockey,⁴⁶ teamgym gymnastics,⁴³ and a variety of sports.⁴² The participants had either acute LBP,⁴³ chronic LBP,^42,46 or a mixture of subacute and chronic LBP.⁴⁵

Outcome Measures Used

For all included studies, outcomes were assessed during and/or after intervention only, between 3 and 12 weeks after baseline.^42,43,46 None of the studies performed long-term follow-up assessments after their postintervention assessments. Four studies used the visual analog scale (VAS) to measure pain intensity;^42,44–46 the remaining study used the Borg Category Ratio Scale.⁴³ One study⁴⁶ used the Oswestry Disability Index (ODI) and short-form 36 questionnaire (SF-36) in addition to the VAS. Three studies used some form of physical function outcome measure.^42,45,46

Intervention Type

The core stability exercises evaluated in the included studies varied. One study used machine-based strengthening exercises for the low back and abdomen,⁴² whereas another used a full-body strengthening program that combined machines, free weights, and body-weight exercises targeting the abdominal and lower back muscles (eg, crunches performed using a Swiss ball, prone “Supermans,” traditional abdominal crunches).⁴⁶ Two of the studies involved exercises where participants cocontracted the transversus abdominis and multifidus muscles and progressed from easier to more complex positioning and the use of unstable surfaces.^43,44 The remaining study⁴⁵ used “dynamic muscular stabilization technique,” which also involves a progression of resistance and more difficult positioning through different stages.

Control/comparison group interventions also varied widely between the studies—for example, therapeutic ultrasound, short-wave diathermy, and floor-based lumbar strengthening exercises (which could also be considered a form of core stability exercise);⁴⁵ Souchard postural exercises with stretching and unweighted abdominal and back muscle strengthening exercises (potentially another form of core stability exercise);⁴² and regular training and recreational activities.^43,46 One observational study did not include a control or comparison intervention.⁴⁴

Primary Study Outcomes: Back Pain and Disability

Table 3 depicts the results of the included studies. Two RCTs (45 participants) and 3 nonrandomized studies (47 participants) provided information on the effect of core stability exercises on pain and disability outcomes. Four of the 5 studies reported a statistically significant improvement in average pain intensity for their core stability intervention groups,^42,44–46 and clinically meaningful differences in average pain intensity were observed in 3 of the 5 studies.^42,44,45 Average pain intensity in both core stability intervention groups in the study by Jackson et al⁴⁶ demonstrated statistically significant but not clinically important improvement over time, whereas the average pain intensity in their control group increased by the end of the trial. Two studies^42,45 reported statistically significant improvement in pain intensity in their comparison treatment groups over time.

TABLE 3: Summary of Results

Two RCTs (75 participants) and 2 nonrandomized trials (69 participants) provided information on the effect of core stability exercises compared with other treatments. One study⁴⁵ noted statistically significant and clinically important improvements in pain intensity, favoring the core stability intervention group over their conventional treatment group (which included lumbar strengthening). Jackson et al⁴⁶ noted statistically significant but not clinically important improvements in pain intensity in both of their core stability intervention groups compared with the control group, with no statistically or clinically significant differences noted between the 2 intervention groups. The remaining studies^42,43 did not find statistically significant or clinically important differences in pain intensity between their intervention and comparison groups.

Jackson et al⁴⁶ found that disability due to LBP (ODI) showed statistically significant and clinically important improvement in both of their intervention groups compared with the control group, with no statistically significant or clinically important differences noted between the 2 intervention groups.

Secondary Study Outcomes: Functional Testing and Global Health Findings

Jackson et al⁴⁶ found statistically significant increases in strength for all of the functional measures assessed (bench press, leg press, lat pulldown) in both of their intervention groups compared with the control group. One study⁴² found statistically significant increases in peak isokinetic torque at 60 degrees per second for the lumbar flexors and extensions in the intervention group, whereas the comparison group demonstrated a statistically significant increase in peak isokinetic torque at both 60 degrees per second and 120 degrees per second. Another study⁴⁵ found statistically significant improvements in all functional measures assessed (walking, stand-ups, and climbing) in their intervention and comparison groups, with statistically significant differences favoring the intervention group.

Jackson et al⁴⁶ further found that quality of life (SF-36) was significantly improved in both of their intervention groups compared with the control group in both the physical and mental composite scores, with no statistically significant differences noted between the 2 intervention groups. None of the included studies reported any adverse events occurring during the trials. However, only one of the studies⁴⁵ specifically indicated an absence of adverse events.

Methodological Quality

One of the non-RCTs⁴³ was found to have moderate quality, and one of the RCTs⁴⁵ was found to have a low risk of bias. The remaining studies could be considered low quality or at a high risk of bias. Downs and Black checklist scores of the 3 non-RCTs ranged from 6 to 11. The critical items missing from all 3 non-RCTs were: adequate descriptions of the representativeness of participants, treatments, and settings; a lack of participant and provider blinding; a description of study compliance or adverse events; and adequate power calculations. The 2 included RCTs had 2⁴⁶ and 7⁴⁵ items with low risk of bias; common potential biases were a lack of blinding of participants and providers and nonreporting of treatment compliance. The overall quality of the available evidence was rated as low, as we judged future research to be highly likely to change our estimate of the effect.

DISCUSSION

We found only a small number of studies focusing on the use of core stability exercise as a treatment for athletes with LBP, including only 2 RCTs. There was important heterogeneity among included studies and an overall low quality of evidence available. Most of the included studies reported statistically significant and clinically important improvements in pain intensity over time in the groups performing core stability exercises. Two RCTs, 1 with a low risk of bias, demonstrated statistically significant and clinically important differences in pain intensity and functional outcomes, favoring core stability exercise over conventional treatment⁴⁵ and regular training.⁴⁶ Differences in the studies included the ages of the athletes involved, the sports in which they participated, and the different forms of “core stability exercise” used as an intervention. This heterogeneity precluded meta-analysis. Furthermore, none of the studies reported long-term follow-up, and sample sizes were small.

Core stability exercises have received increasing scrutiny recently.^29,47 One survey indicated that core stability exercises are the most frequently recommended form of exercise by Irish physical therapists.⁴⁸ Previous systematic reviews and meta-analyses have demonstrated that core stability exercise programs can be effective for specific populations with LBP, particularly those with chronic LBP.^21,23,24,30 The limited findings provided by the current review of athletes with LBP undergoing core stability exercise programs demonstrates conflicting findings in both within-group and between-group differences, particularly when considering clinically important differences in pain intensity and functional outcomes. That being said, the 3 studies that included athletes with chronic LBP^42,45,46 generally reported statistically significant and clinically important within-group and/or between-group improvements in pain intensity for their core stability groups. However, only one of those studies⁴⁵ on patients with chronic LBP was found to have a low risk of bias. The one study that specifically included athletes with acute LBP⁴³ did not report statistically significant improvements in pain intensity over time for their core stability group. These findings are similar to what has been found for the general population.

Issues with muscular weakness, imbalance, and recruitment specific to the hip and/or core musculature have been implicated among many potential sources of LBP in athletes.^49–53 Athletes with a history of LBP are also at increased risk for future LBP.⁵⁴ Therefore, it seems plausible that core stability exercises would be a potentially suitable intervention to consider for athletes with LBP. Although core stability exercises may be useful in treating athletes with chronic LBP, it cannot be conclusively stated which form(s) of core stability exercise are most effective because the included studies varied.

Strengths and Weaknesses

Among the strengths of this review were the searching of multiple databases and using reference searching, along with authors independently evaluating database search results and assessing the literature for quality and risk of bias, with a third author resolving differences. However, the relatively small number of studies that met the inclusion criteria and the overall low quality of included studies could lead to bias in the conclusions. In addition, the inclusion of nonrandomized comparative studies and uncontrolled studies in this review could potentially be a source of bias. Another limitation was that only articles in English were retrieved. However, as the search strategy was conducted in all languages, such cases should have been identified. It is also possible that there is publication bias with respect to the use of core stability exercises in athletes with LBP, particularly given the popularity of core stability exercises in the lay media and their use among the general population.

The heterogeneity of the included studies (eg, athlete characteristics, nature of the “core stability exercise” interventions) limits the interpretation of the findings of this review and their applicability in clinical practice. It is also acknowledged that some participants in studies evaluating core stability exercises for LBP in the general population may be athletes from different sports and at different levels of competition. However, such studies also include nonathletes and do not specifically analyze the results of athletes versus nonathletes.

Future Research Directions

Further research is clearly needed on this topic. Researchers and clinicians need to determine whether core stability exercises should be used to treat low back injuries, prevent low back injuries and possible recurrences, or if they should be used for performance enhancement. Perhaps, the role of core stability exercises is a combination of the above. A consistent definition of what comprises a core stability exercise must be used in future research. This could be determined through a consensus study.

Higher-quality adequately powered RCTs are needed to determine which interventions have the greatest effect on athletes with LBP and should compare different forms of core stability exercise, both alone and in combination with other treatments (eg, manual therapies, biopsychosocial interventions, different forms of exercise therapy). The effects of core stability exercises in athletes with LBP of different durations (ie, acute, subacute, or chronic) across different age groups (youths, adults, seniors), from different sports, and at different levels of competition (recreational, amateur, professional), each need to be further evaluated. Continued use of the VAS as an outcome measure is advocated for future research, as well as some form of functional outcome, and a measure of disability due to back pain (eg, ODI). The use of standardized exercise protocols and outcome measurement in future research would be beneficial in allowing for meta-analyses to be performed.

CONCLUSIONS

The quantity and quality of the evidence available on the use of core stability exercise as an intervention for athletes with LBP is low. The types of exercises evaluated in the included studies varied considerably as did study populations, which were small and saw only mixed results in terms of short-term pain intensity and disability. Pooling of data was not possible and the state of the evidence precludes conclusions from being drawn at this point. Further research on this topic is clearly needed and should include consistent definitions of both study populations and interventions.

ACKNOWLEDGMENTS

The authors thank and acknowledge Anne Taylor-Vaisey of the Canadian Memorial Chiropractic College and Mary Chipanshi of the University of Regina for their assistance with formulating the search strategy and conducting the literature search.