The effectiveness of different aerobic exercises to improve pain intensity and disability in chronic low back pain patients: a systematic review

Background: Physical activity, including aerobic exercise, is highly recommended for chronic low back pain (CLBP) patients to improve pain intensity and functional disability. Objectives: To assess the effectiveness of different aerobic exercises to reduce pain intensity and functional disability in patients with CLBP. Methods: A computer-aided search was performed to find Randomised Controlled Trials (RCTs) that evaluated the effectiveness of different aerobic exercises in CLBP. Articles published between January 2007 to December 2020 were included in the review. Quality assessment using the PEDro scale, extraction of relevant information, and evaluation of outcomes were done by two reviewers independently. Results: A total of 17 studies were included that involved 1146 participants. Outcomes suggested that aerobic exercise combined with other interventions was more effective than aerobic exercise alone. Aerobic exercise with higher frequency (≥ 5 days/week) and longer duration (≥ 12 weeks) were effective to gain clinically significant (≥ 30%) improvements. Environment and using pedometer did not seem to influence the outcomes. Conclusions: Pain intensity and functional disability in CLBP patients can be minimized by prescribing aerobic exercise. However, to get better improvements, aerobic exercise should be done in combination with other interventions and at optimum frequency and duration. Further studies should emphasize examining the optimal doses and duration of different aerobic exercises.


Introduction
Low back pain (LBP) is one of the leading causes of disability-related musculoskeletal conditions globally. 1,2It is reported that 70-80% of the population suffer from LBP at some point in their lifetime. 2,36][7] The National Health Service (NHS) spends more than £9 billion to provide CLBP patients treatment. 8CLBP is now regarded as a significant public health problem globally, and the prevalence of CLBP has risen noticeably in the past decades. 9ronic low back pain (CLBP) is defined as pain, muscle tension, or stiffness located between the lower rib margins and above the lower gluteal folds that persists for more than 12 weeks (three months) with or without symptoms in the lower limbs. 10Decreased physical activity is regarded as one of the main contributing factors to chronic musculoskeletal pain Criteria for considering studies for this review Types of studies: In this systematic review, we included randomized controlled trials (RCTs) that evaluated the effectiveness of AE with or without other interventions in at least one group.We only considered the studies that were published in the English language.The study that investigated patients with chronic low back pain and acute and/or subacute LBP together was also excluded.Papers published before 2007 and other than the English language were excluded.

Types of participants:
We included the studies involving participants aged ≥18 years with low back pain for a minimum of 3 months (≥12 weeks).Studies that involved patients with a history of acute or subacute low back pain, cauda equine syndrome, inflammatory or tumoral back conditions, osteoporosis of the spine and pregnancy, and surgery in the lumbosacral region, spinal fracture, and dislocation were excluded in this systematic review.

Types of outcome measures:
Studies that evaluated at least pain intensity or functional disability as the outcome with or without other measurements were included in this systematic review.
Details of inclusion and exclusion criteria are shown in Table 1.

PubMed
The following keywords were used independently and in combination: low back pain, backache, aerobic exercise, walking, treadmill walking, cycling.

Types of studies
• Used a randomized controlled trial design.• Examined aerobic exercise with or without other intervention in at least one group.

• Papers published between
January 2007 to December 2020.
• The study that investigated patients with chronic low back pain and acute and/or subacute LBP together.• Papers published other than the English language.

Types of participants
• Aged ≥ 18 years • Patients with low back pain for a minimum of 3 months (≥12 weeks).
• Patients with acute or subacute low back pain.
• Patients with cauda equine syndrome.
• Patients with inflammatory or tumoral back conditions.• Patients with osteoporosis of spine and pregnancy.• Patients with a history of surgery in the lumbosacral region, spinal fracture, and dislocation.

Types of outcome measures
• Studies that evaluated at least pain intensity or functional disability as the outcome with or without other measurements.

Assessment of risk of bias and quality of studies
The Physiotherapy Evidence Database (PEDro) scale was employed to assess the methodological quality and risk of bias of included studies.PEDro scale is regarded as a valid and reliable risk of bias tool. 35,36PEDro scale has 11 components or items including eligibility criteria, random allocation, concealed allocation, baseline similarity, blind subjects, blind therapists, blind assessors, sufficient follow up (85% follow up for at least one key outcome), intention-to-treat analysis, between-group statistic comparison (for at least one key outcome), and point estimates and variability (for at least one key outcome). 37Eight items (item 2-9) are used to evaluate the risk of bias and last two items (10 and 11) are related to statistical reporting. 36The first item which is eligibility criteria is not counted in the total score as it is related to external validity.Hence, PEDro score ranges between 0 and 10 points, where the article with higher score regarded as better article in terms of risk of bias and statistical reporting.Any study with a score between 6 and 10, score with 4 or 5, and score ≤ 3 is considered good quality, fair quality, and poor quality study, respectively. 38,39However, it is impossible to blind therapists and all subjects in clinical trials because of ethical standards.Hence any study with a score of 8/10 is regarded as low risk of bias. 39PEDro scoring was done using the PEDro scale by the first two authors independently, and the other two authors resolved any discrepancy through discussion.

Types of outcome measures
In this systematic review, we evaluated the effectiveness of different AE in improving pain intensity and disability in CLBP patients.Therefore, we reported the following outcomes: • Pain intensity: Pain intensity measured by a pain scale, including Visual Analogue Scale (VAS), Numerical Pain Rating Scale (NRS), and McGill Pain Questionnaire (MPQ).
Outcome measurements of different time points were included in the analysis to evaluate the treatment effect described below.

Data synthesis and measurement of treatment effect
Required data were extracted from studies by using a data extraction form.The first two reviewers extracted relevant information on sample size and subject characteristics; type, frequency, intensity, and duration of interventions; instruments used to assess the outcomes; and outcomes of pain intensity and functional disability.The third and fourth authors further evaluated the extracted data, and any disagreements were resolved through discussion.The outcomes were continuous variables, and treatment effects were reported as mean differences and mean percentage changes.1][42] We did not perform the meta-analysis due to the heterogenicity of participants, intervention, and outcome measures.

Data search
After searching databases, a total of 1,145 studies were identified.After removing duplicates, 1135 articles were screened by title and content of the abstract.After that screening, 1055 articles were excluded, and the remaining 80 articles were evaluated for eligibility.Finally, a total of 17 studies met all inclusion criteria and were included for this review.
A PRISMA flow diagram of the study selection process is shown in Figure 1.

Study characteristics
All included studies were RCTs, involved a total of 1146 patients.The number of participants ranging from 14 43 to 246, 44 and patients were >18 years.The duration of included studies' intervention was between 4 weeks 45 and 12 months. 46mong these, eleven studies followed up for short term (<3 months or 12 weeks), 45,[47][48][49][50][51][52][53][54][55][56] eleven studies followed up for intermediate-term (3 months to <12 months) 43,44,[46][47][48]52,53,[56][57][58][59] and two studies followed up for long term (≥12 months). 46,55 Studis examined the effectiveness of different types of AE. Eleve studies evaluated the effectiveness of different forms of AE alone [43][44][45][46][47]49,54,[56][57][58] while seven studies examined AE in conjunction with other intervention including stabilization exercise, 47 back school program, 50 rehabilitation program, 51 group exercise class, 52 traditional physiotherapy 55 and home exercise.48 As an intervention, ten studies used walking exercise [45][46][47][48][49]51,53,54,58 ; three studies used stationary cycling exercise 45,50,52 ; one study used treadmill running exercise 57 ; one study used both walking and running exercise 43 ; one study used combined treadmill walking, stair climbing and stationary cycling exercise 59 ; one study used walking and jogging exercise with elliptical trainer 56 ; one study used individually designed and supervised aerobic exercise.55 In this review, we focused on the improvement in pain intensity and functional disability.To measure pain intensity, nine studies used the Visual Analogic Scale (VAS), 43,[47][48][49]51,52,55,56,59 six studies used Numerical Pain Rating Scale (NRS), [44][45][46]50,53,58 one study used McGill Pain Questionnaire (MPQ) 57 and one study did not measure pain intensity. 54 Functionl disability was evaluated by using Oswestry Low Back Pain Disability Index (ODI, 0-100) in eleven studies, [43][44][45]47,49,[52][53][54]56,58,59 Oswestry Low Back Pain Disability Index (ODI,10-60) in one study, 51 Roland and Morris Disability Questionnaire (RMDQ) in four studies 46,48,50,57 and Aberdeen Low Back Pain Disability Scale (ALBPS) in one study.55    Risk of bias and quality assessment of studies Quality assessment of included studies using the PEDro scale is shown in Table 2, which demonstrated that the range of the scores was between 4/10 and 8/10 (mean 6. 35 AE 1.46).46,47,49,[52][53][54][55]58,59 Studies that scored the lowest had a lack of concealed allocation, blind assessors, adequate follow-up, and intention-to-treat analysis.45,51 Effectiveness of different aerobic exercises to improve CLBP Walking alone exercise Bello et al. 50ompared walking exercise alone to lumbar stabilization exercise by involving a total of 50 patients who were divided into walking exercise group (WG) and lumbar stabilization exercise group (LSG).WG received walking exercise on the treadmill at an intensity of 65-80% HRR, while LSG received 30 mins of lumbar stabilization exercises following the McGill protocol for eight weeks (three times/week).50 After intervention, both groups showed improvement in pain (WG vs LSG = 32.8% vs 59.4%) and disability (WG s LSG = 14.4% vs 48.9%), while LSG demonstrated better outcomes. 50In another study, Shnayderman and Katz-Leurer 54 evaluated the effectiveness of walking exercise (WG) against specific low back strengthening exercises (SG) by recruiting 52 patients.WG received 40 minutes of walking exercise on the treadmill at 50% heart rate reserve. 54Both groups received exercises two days a week for six weeks.54 Outcomes revealed that both groups showed significant improvements in disability without significant differences between groups where ODI scores were reduced by 34.3% in WG and 30.6% in SG. 54 Both studies' significant limitations included lack of a control group, no long-term follow-up, and a short intervention period. Coversely, Hurley et al. 44 evaluated walking exercise's effectiveness with a larger sample size and long-term follow-up. 24atients aged 18-65 were equally divided into three groups.The first group (WG) received supervised walking exercise for a minimum of 10-minutes to 30 minutes walk/day at 40-60% HRR, at least four times/week for seven weeks.44 The second group (ECG) trained with group exercise class (Back to fitness program, a one-hour long class per week for eight weeks) and exercise including warm-up and stretching.The third group (UG) received usual physiotherapy.44 Results showed that pain and disability improved in all three groups.44 Authors reported that 48% in WG, 45% in ECG, and 31% in UG participants achieved minimal clinically significant difference (MCID) in the ODI score.44 Whereas 44%, 29%, and 37% of WG, ECG, and UG participants reached MCID in the NRS score.44 Authors also reported that the walking program had the greatest adherence and the lowest costs. 44 A significant limitation of thisstudy was that a total of 40 therapists were involved in this study to train the patients; hence therapist effects could influence the outcomes.

Walking exercise in conjunction with other intervention
Cho et al. 51 studied whether treadmill walking exercise combined with a low back pain rehabilitation program helped reduce pain and disability in CLBP patients.Twenty men were equally divided into an experimental group (EG) and a control group (CG). 51EG received treadmill walking exercise without a slope at 3-3.5 km/h, for 30 minutess and low back pain rehabilitation program; whereas, CG received only a low back pain rehabilitation program. 51Both groups received 30 minutes long low back pain rehabilitation program, three days/week, and the duration of intervention was eight weeks. 51After the intervention, both groups showed improvement in pain (VAS) and disability (ODI) scores without any significant difference between groups. 51In EG, VAS and ODI scores were reduced by 46.1% (vs 43.5% in CG) and 21.5% (vs 12.7% in CG) respectively.Overall, additional treadmill exercise did not provide additional improvements. 51he small sample size was a major drawback of this study.
Koldas et al. 48experimented with a larger sample of sixty patients to examine combined walking exercise and home exercise effectiveness.Twenty patients (AHE) received 40-50 minutes of exercise on a treadmill at 65-70% HRR, three times/week with home exercise. 48At the same time, the remaining 40 patients were assigned to receive either physical therapy (PT) or home exercise only (HE). 48Home exercises included basic flexion, extension, mobilization, and stretching, and the patients were asked to perform the exercise once a day with 15-20 repetitions. 48All groups received their specific exercises for six weeks. 48Results showed that pain reduced significantly in all three groups after the treatment (AHE vs. PT vs. HE: 39.6% vs. 36.5% vs. 28.6%)and at one-month follow-up (AHE vs. PT vs. HE: 38.8% vs. 53% vs. 40%). 48Disability was improved significantly in AHE and PT at both post-intervention (25.2% vs. 25.2%) and follow-up (22.7% vs. 30.3%),while in HE, it was negligible. 48n another study, Suh et al. 47 utilized 48 patients aged > 20 years to compare walking exercise alone (WE) to three different interventions, including combined walking and stabilization exercise (SWE), flexibility exercise (FE), and only stabilization exercise (SE).The WE group received 30 mins of fast walking exercise on flat ground with abdominal bracing, whereas the SWE group trained with 30 minutes of walking exercise and 30 minutes of stabilization exercise. 47utcomes were measured at baseline, within two weeks after intervention and six weeks after the intervention. 47Results indicated that pain intensity decreased in all four groups both during activity (FE vs WE vs SE vs SWE: 45.26% vs 38.7% vs 48.19% vs 44.06%) and at rest (FE vs WE vs SE vs SWE: 33.61% vs 18.25% vs 35.33% vs 38.9%) after the intervention. 47Further assessment at six weeks after intervention showed that all groups retained enhancement in pain scores. 47Disability evaluation demonstrated that after the intervention, the ODI score was decreased by 19.93% in WE; while in FE, SE and SWE, it was 16.05%, 19.43%, and 18.09%, respectively. 47The frequency of exercise in SE and WE increased significantly after the intervention.However, the SWE group showed the opposite trend, which demonstrated poor adherence to exercise, and it was difficult for the participants of the SWE group to perform 60 minutes of exercise. 47dometer-driven walking exercise Eadie et al. 58 experimented on 60 patients aged 18-70 years by distributing them into three groups (WG, SG, PG).WG received walking exercise for 30 minutes at moderate intensity, five days/week, and they were asked to wear a pedometer during walking to record the progress. 58SG Received a supervised exercise class (back to fitness program) once per week, whereas PG received usual physiotherapy. 58The total intervention duration was eight weeks, and outcomes were measured at three months and six months. 58Pain score evaluation showed that both WG (12.1%) and SG (12.9%) gained similar improvements, whereas PG (32%) obtained more significant improvements. 58However, unlike the other two groups, WG (-1.79%) failed to retain the improvements at six months. 58Moreover, the smallest improvement in disability also was in WG (9.4%) compared to SG (22%) and PG (27.3%). 58The authors reported small sample size and a high drop-out rate during the follow-up period could impact the outcomes. 58sides, McDonough et al. 53 examined 57 patients aged between 43 and 53 years.The experimental group (EG) received combined pedometer-driven walking exercise and education, whereas the control group (CG) received only education or advice.Participants were familiarized with wearing a pedometer, and they were asked to record their daily steps in a walking diary. 53The intervention duration was nine weeks, and measurements were done after intervention and six months after randomization. 53Results indicated pain intensity improved in both groups; however, EG showed greater improvement (16.7% vs. 15.2% at nine weeks and 29.6% vs. 10.9% at six months). 53EG showed a better outcome in disability (17.2% vs 3.3% at nine weeks and 25.7% vs 5.5% at six months). 53However, the sample size was relatively small, which was a weakness of this study.
In contrast, Krein et al. 46 experimented with a larger sample size and long-term follow-up to examine whether additional support affected patients' improvement.They examined 229 patients by separating them into two groups: experimental group (EG) and usual care group or control group (CG). 46EG received an uploading pedometer and additional support, access to a website that provided information about walking goal progress, and patients received feedback, motivational and informational messages. 46In contrast, CG received an uploading pedometer but did not receive any walking goal and did not access the website. 46This study's duration was 12 months, and outcome measurements were done at baseline, six months, and 12 months. 46Results demonstrated pain intensity improved in both groups at six months (EG vs CG = 21.7% vs 14.8%) and 12 months (EG vs CG = 10% vs 8.2%), where improvements were greater at six months. 46Disability improvement was also greater in EG at six months (20.9% vs. 6.1%). 46Patients were recruited from one medical center, which was indicated as a limitation of this study. 46

Stationary cycling exercise
Barni et al. 50evaluated the effectiveness of a combined back school program and stationary cycling exercise by recruiting 22 patients.Patients were assigned to either the experimental group (EG) who received exercise on the stationary bike at 65% HRR and back school program; or the control group (CG) who received only the back school program. 50The interventions' total duration was five weeks (90 minutes session, two sessions/week). 50Post-intervention measurements showed a greater reduction in NRS and RMDQ index in the experimental group than in the control group (NRS = 27% vs 13.14% and RMDQ = 25.58%vs 10.3%). 50This study's major limitations were lack of long-term follow-up, short duration of intervention, and a small number of participants.
Marshall et al. 52 experimented with a relatively larger sample size (64 patients) and a longer duration to observe whether stationary cycling exercise combined with exercise class (CEG) was more effective than specific trunk exercise conjunction with exercise class (SEG).Both groups received their specific exercise for 35-40 minutes and 50-60 minutes of exercise classes (three sessions/week) for eight weeks. 52Outcomes were recorded at baseline, post-intervention, and six months from the start of the intervention. 52Results showed that pain decreased in both groups (SEG vs CEG = after intervention: 52.8% vs 17.8%, at six months: 44.4% vs 26.7%), where SEG showed better improvements. 5256% of SEG and 50% of CEG participants showed clinically relevant changes (≥ 30%) after intervention. 52Disability was significantly lower in SEG compared to CEG after the intervention (40.9% vs. 16.3%), and 66% of SEG and 44% of CEG participants demonstrated clinically significant change (≥30%) in ODI score. 52The authors concluded that both exercises effectively improved pain and disability without significant differences between groups. 52Although trunk exercise showed better improvements than stationary cycling immediately after the intervention, long-term follow-up outcomes were similar. 52ulliyil et al. 45 examined whether stationary cycling exercise is superior to treadmill walking to improve CLBP.A total of 30 patients aged 18-30 years were divided into two groups. 45One group (TG) received AE by treadmill walking, while another group (SCG) received AE by stationary cycling. 45Both groups received AE at moderate intensity (13-14 RPE) for 10-20 minutes for four weeks (five days/week). 45Post-intervention measurements demonstrated that both groups showed significant improvements in all measurements without any significant differences between groups (TG vs SCG = NRS at rest: 80.1% vs 69.3%, NRS on activity: 55.4% vs 45.3%, ODI: 53.8% vs 49.8%). 45However, the very short duration of the study and the absence of long-term follow-up were the major drawbacks of this study.

Treadmill running exercise
Chatzitheodorou et al. 57 recruited 20 patients to examine the effectiveness of running exercise to improve CLBP.
Ten patients in the experimental group (EG) received high-intensity AE by running on the treadmill at 60%-85% of HRR, 30-50 minutes session, three sessions/week. 57Whereas the remaining ten patients in the control group (CG) received passive modalities (45 minute session) without any physical activity. 57After intervention (12 weeks) EG showed significantly better improvement in pain (40.1% vs. 0.6%) and disability (30.4% vs. 0.7%) compared to CG. 57

Combination of different aerobic exercises
Murtezani et al. 59 examined 101 patients with CLBP by assigning them either to the experimental group (EG) or the Control group (CG).EG received high-intensity AE, including treadmill walking, stair climbing, and stationary bicycling at 50%-85% HRR, 30-50 mins sessions, three sessions/week for 12 weeks. 59In contrast, CG received passive modalities (45 min session, three times/week). 59Results showed significant improvements in all parameters in EG (pain 66.7% and disability 49%), while the improvements in CG were non-significant (pain 1.6%, disability 0.3%). 59dividually designed and supervised aerobic exercise Chan et al. 55 utilized 46 patients who were included either in the experimental group (EG) or control group (CG).EG received conventional physiotherapy and 20 minutes of aerobic training at 40%-60% HRR, gradually progressed up to 85% at a 5% increment each weak for eight weeks (three sessions per week). 55Subjects in EG were also asked to perform a minimum of one additional training per week at home. 55Participants selected the type of that training according to their preference, including treadmill walking or running, stepping, and cycling. 55Subjects in the control group received only conventional physiotherapy for eight weeks. 55Post-intervention measurements indicated significant improvements in pain and disability in both groups without any significant differences between groups. 55EG attained clinically significant improvements (≥ 30%) in VAS and ALBPS scores at all time points. 55The short duration of intervention and relatively small sample size are major limitations of this study. 55In addition, the authors reported poor baseline fitness level of patients, which could be a factor to influence the outcomes. 55fluence of the environment Kanitz et al. 43 experimented on 14 patients to see the impact of the environment on AE outcomes.Participants were randomly allocated into two groups and received 35 minutes of walking/running exercise at moderate intensity (85-95% HRvt2) for 12 weeks (two times/week) either on land (LG) or in water (AG). 43Outcomes showed improvements in pain and disability score in both groups without any difference between groups (LG vs. AG = pain: 66.67% vs. 47.27% and disability: 40.59% vs. 48%). 43However, small sample size and absence of long-term follow-up were reported as weaknesses of this study. 43able 3. Overview of included studies that evaluated the effectiveness of aerobic exercise alone intervention.Kanitz et al. 43 14 patients, Aged: 30-50 years
• • CG: Received only education or advice.

Periodized progressive overload training
Kell and Asmundson 56 carried out an experiment to observe periodized progressive overload training effectiveness.A total of 27 patients with CLBP were equally divided into three groups (AT, RT, and CG). 56AT received periodized progressive overload aerobic training (elliptical trainer and treadmill walking and jogging.20-35 min session, three sessions/week), while RT received periodized resistance training. 56Patients in the control group (CG) maintained regular activity. 56The intervention's total duration was 14 weeks, consisting of two phases (seven weeks per phase). 56Outcomes were measured after completing each phase (at eight weeks and 16 weeks). 56Results showed significant improvement in pain (27.8% at eight weeks and 38.9% at 16 weeks) and disability (30.2% at 8 weeks and 40.1% at 16 weeks) in RT.In contrast, AT demonstrated improvement only in cardiorespiratory performance (VO2max), body fat, and body mass. 56owever, an experiment with a larger sample size is needed to conclude the efficacy of periodized progressive overload aerobic training on CLBP patients.
The summary of interventions, measurements, outcomes, and main limitations of the included studies are shown in Table 3 and Table 4.

Discussion
Walking is a highly cost-effective AE that is regularly advised to patients with CLBP. 44,60,61This exercise is easy to perform and does not require any particular skill or facilities. 60,62It is regarded as one of the safest exercises because of its low injury rate. 60,62Also, walking enhances cardio-respiratory capacity, maximum oxygen uptake, and prevents LBP by increasing the isometric endurance of muscles. 63,64 this review, three good-quality studies (PEDro score ≥ 7) compared walking alone exercise to other interventions. 44,49,54Studies showed that walking alone exercise effectively reduced pain and disability in CLBP patients. 44,49,54owever, there was no evidence that walking alone exercise was superior to other interventions.][32] Moreover, in this review, walking alone exercise was not effective in attaining clinically significant improvements (≥ 30%) in pain intensity and disability on most occasions. 44,49,54In previous studies, exercise alone therapy was not effective to make clinically significant changes (≥ 30%) in CLBP patients; therefore, it was suggested to apply combined treatment in clinical trials. 22,65Lawford et al. 32 also asserted in their review that combined walking exercise was more effective than walking exercise alone.
In this review, three studies (fair to good quality) examined the effectiveness of walking exercise combined with other interventions. 47,48,51Results indicated that walking exercise combined with other interventions effectively reduced pain intensity in CLBP patients, and the improvements were clinically meaningful (≥30%). 47,48,51Besides, disability improvements were statistically significant in all three studies. 47,48,51However, these enhancements (<30%) were not remarkable enough to be clinically significant.It was claimed that to get better improvements in any intervention, patients should be trained for a sufficiently longer period. 51The duration of these studies was ≤ 8 weeks; hence, the duration could be too small to make any clinically significant improvement in disability.Moreover, exercise effectiveness can be improved by adjusting the intensity and duration of exercise according to patients' capacity. 66Research showed that atrophic changes in lumbar paraspinal muscles are common in CLBP patients, which could decrease patients' ability to do prolonged exercise. 67,68One of the included studies also showed that it was difficult for the patients to continue 60 minutes of exercise. 47Hence the authors advised selecting an exercise program with a duration of about 30 minutes. 470][71] Some studies were carried out to see if pedometer-driven walking exercise could benefit CLBP patients' conditions. 46,53,58In this review, the findings of three included RCTs (PEDro score = 8) indicated that walking exercise in conjunction with advice, education, or support effectively improved CLBP patients' conditions. 46,53,58However, there was no clear evidence that using a pedometer had any extra benefits.A pedometer can guide a patient to track their progress, but further research is needed to recommend pedometer as rehabilitation tools for CLBP patients.
Stationary cycling exercise is another form of AE which is regarded as one of the most effective exercises to improve muscular coordination. 50,72Three fair to good quality RCTs were included in this review that examined the effectiveness of stationary cycling exercise, and outcomes demonstrated that stationary cycling exercise was as effective as walking exercise, and stationary cycling exercise was not inferior to other interventions. 45,50,52In addition, Chatzitheodorou et al. 56 showed that high-intensity AE, including running, was adequate to improve pain and disability in CLBP patients significantly.In another study, Murtezani et al. 59 demonstrated that a combination of different AE effectively improved conditions of CLBP patients.These included studies also showed the possible influence of exercise frequency and duration of intervention on outcomes.The study with the higher frequency of exercise (five days/week) 50 and longer duration (12 weeks) of intervention 57,59 showed clinically significant (≥ 30%) changes in both pain and disability scores. 504][75] Therefore, the exercise frequency and duration of intervention could be the keys to obtain clinically significant improvements in pain intensity and disability in CLBP patients.However, future studies with a larger sample size and long-term follow-up are required to justify it.
Individually tailored and supervised exercise programs were suggested by Hayden et al. 76 in their meta-analysis.In this review, one good quality study (PEDro = 7) showed that individually designed and supervised aerobic exercise effectively made clinically significant improvements in patients with CLBP. 55In addition, one good quality study (PEDro = 7) was included in this review that examined the impact of the environment on the effectiveness of aerobic exercise. 43Patients were randomly allocated and received the same exercise either on land or in water, and results indicated that the environment did not influence the outcomes. 43However, more studies with a large sample size and long-term follow-up are required for further evidence.
Only one included study in this review did an experiment to observe the effectiveness of periodized progressive overload training. 56Results showed that periodized progressive overload aerobic training failed to improve pain intensity and disability in CLBP patients. 56However, the study was regarded as fair quality (PEDro score 5); therefore, a good quality study with a larger sample size is warranted to conclude the effectiveness of periodized progressive overload aerobic training on CLBP patients.

Strengths and limitations of the review
There are several strengths of this review.A highly sensitive search of different databases was performed to find the relevant studies.A total of 17 RCTs were included in this review that involved an adequate sample size of 1146 participants after applying strict inclusion and exclusion criteria.Moreover, PEDro scoring was done to assess quality and risk of bias, and all included studies scored ≥ 4 to be considered fair to good quality.
In contrast, in this review, only improvements in pain intensity and disability were evaluated, which was considered a major limitation.Other outcomes, including fear-avoidance beliefs, mental and physical health, quality of life, and costeffectiveness of interventions, were missed.In addition, studies published other than the English language were excluded in this review.Therefore, it is highly recommended for further review to include studies in different languages and from lesser-known databases.

Conclusion
Overall, this review showed that AE effectively reduced pain intensity and functional disability in CLBP patients.It also demonstrates the appositeness of using AE as an intervention in future studies.Findings of most of the included studies demonstrated that patients gained statistically significant pain intensity and disability improvements.Results indicated that exercise should be done under supervision at a minimum frequency of 5 days/week, for at least 12 weeks, and in combination with other interventions including education, physiotherapy, home exercise, or other forms of exercise to get clinically significant outcomes.Future studies should emphasize training patients at the optimum frequency, intensity, and duration so that the participants can achieve clinically meaningful improvements.

Data availability Underlying data
All data underlying the results are available as part of the article and no additional source data are required.

Reporting guidelines
Mendeley Data: PRISMA checklist for 'The effectiveness of different aerobic exercises to improve pain intensity and disability in chronic low back pain patients: A systemic review.' https://doi.org/10.17632/79vnhtfh85.2 77 This project contains the following data: intervention with which the interventions (aerobic exercise) were compared.The agent of comparison is not specified.However, it has to be reminded here that the study hypothesis is different.
The contents of Table 1 are already elaborated upon in the text.This must be removed.Likewise, since the PEDro database is well-known, the authors may wish to remove the section describing its components.

7.
What heterogeneity was observed in the measurement of outcomes?8.
Due to the fact that the control groups were given distinct exercises and the study's objectives were not met, the results cannot be considered valid.The ideal study would have broken the results of aerobic exercises into modes (different categories of exercises), durations (short, medium, and long), intensities (moderate or high).

9.
The results do not support the study's conclusions.10.It is an update to previous systematic reviews but I have some concerns before approval:

Are
1.The search query is not disclosed; it should at least be contained as a supplement.
2. The reason for not doing a meta-analysis is not described, is this due to issues of heterogeneity?
3. The main outcome measure is not described in terms of the questionnaire.Reviewer Expertise: Low back pain I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.
The benefits of publishing with F1000Research: Your article is published within days, with no editorial bias • You can publish traditional articles, null/negative results, case reports, data notes and more • The peer review process is transparent and collaborative • Your article is indexed in PubMed after passing peer review • Dedicated customer support at every stage • For pre-submission enquiries, contact research@f1000.com
PRISMA checklist.docData are available under the terms of the Creative Commons Attribution 4.0 International license (CC-BY 4.0).

1 Reviewer
the rationale for, and objectives of, the Systematic Review clearly stated?NoAre sufficient details of the methods and analysis provided to allow replication by others?NoIs the statistical analysis and its interpretation appropriate?Not applicableAre the conclusions drawn adequately supported by the results presented in the review?NoCompeting Interests: No competing interests were disclosed.Reviewer Expertise: painI confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptablescientific standard, however I have significant reservations, as outlined above.Version Report 11 March 2022 https://doi.org/10.5256/f1000research.79315.r126016Navid Moghadam Sports Medicine Research Center, Neuroscience Institute, Tehran University of Medical Sciences, Tehran, Iran Thanks for the opportunity to review.
Are the rationale for, and objectives of, the Systematic Review clearly stated?YesAre sufficient details of the methods and analysis provided to allow replication by others?PartlyIs the statistical analysis and its interpretation appropriate?PartlyAre the conclusions drawn adequately supported by the results presented in the review?PartlyCompeting Interests: No competing interests were disclosed.

Table 2 .
Quality assessment of included studies using PEDro scale.

Table 4 .
Overview of included studies that evaluated the effectiveness of aerobic exercise combined with other intervention.