Effectiveness of backward gait training on body composition, cardiopulmonary fitness, inflammation, and metabolic marker in adults: A systematic review and meta-analysis

ABSTRACT

Walking is a fundamental physical activity with significant health implications. Backward gait training (BGT) has emerged as a novel approach with potential benefits, yet its effects in comparison to traditional forward gait training (FGT) remain uncertain. This systematic review and meta-analysis aimed to evaluate the effects of BGT on body composition, cardiopulmonary fitness, and inflammatory and metabolic markers in adults. A comprehensive search across electronic databases was conducted following the Preferred Publishing Items for Systematic Reviews and Meta-Analyses guidelines. Randomized clinical trials (RCTs) comparing BGT with FGT in adults were included. Methodological quality was assessed using the Cochrane risk-of-bias tool. The certainty of evidence was evaluated using the Grading of Recommendation, Assessment, Development, and Evaluation approach. The analysis included a total of 379 male participants across the studies. The meta-analysis demonstrated significant changes in body composition and inflammatory marker outcomes, which included waist-to-height ratio (standardized mean difference [SMD]-1.18, 95% confidence interval [CI]-1.89–0.48, I2 = 83%, P < 0.01), body mass index (SMD-0.55, 95% CI-0.77–0.32, I2= 0%, P < 0.01), and C-reactive protein (SMD-0.98, 95% CI–1.28-0.70, I2= 0%, P < 0.01). In addition, the qualitative review revealed potential enhancements in cardiopulmonary fitness and metabolic markers following BGT. While the results suggest potential benefits of BGT on body composition and inflammatory markers, the evidence remains limited and heterogeneous. Further robust research with diverse populations, longer intervention periods, and comprehensive outcome assessments is essential to elucidate the true impact of BGT and its utility for promoting overall health and well-being in adults.

INTRODUCTION

Gait, commonly known as walking, is a remarkable interplay of neurophysiological processes and biomechanical principles.[1] The act of walking necessitates the precise coordination of multiple muscle groups, facilitating a forward progression of the body while ensuring balance and energy efficiency.[2,3] Numerous studies have provided evidence for the effectiveness of walking on substantial health outcomes, ranging from body fat reduction and amelioration in resting blood pressure to a significant decrease in all-cause mortality risk.[4,5,6,7] Recognizing these benefits, both the Physical Activity Guidelines for Americans and the Centre for Disease Control and Prevention have recommended adults engage in walking for disease prevention, improved health outcomes, and overall well-being.[8,9]

Parallel to the vast research on forward gait training (FGT), backward gait training (BGT) is an emergent area of study, primarily within the rehabilitation domain.[10,11] Contrary to perceptions, backward gait is not a mere reversal of forward motion.[12] Certainly, while neural controls might be congruent,[13] the biomechanics differ substantially. Kinematic studies elucidate the diminished ankle plantar flexion and augmented dorsiflexion during BGT, with a distinct plantar pressure distribution pattern.[14,15] Electromyographic investigations further explain the differential muscular activation patterns between the two gait modalities.[16,17] The backward gait appears similar to forward gait but in reverse direction.[18] Previous study has demonstrated that both forward walking and retro walking exhibit similar neural control and exercise adaption responses.[13]

Backward gait is not just a simple reverse motion of forward gait; it poses distinct biomechanical and neuromuscular demands on the body.[19] From a cardiopulmonary perspective, BGT may increase cardiovascular demand due to the unfamiliar nature of the activity, leading to an elevated heart rate and increased respiratory rate as compared to traditional forward walking at similar speeds.[20,21] This heightened cardiovascular response, over time, could potentially stimulate cardiovascular endurance.[22] On the metabolic front, the biomechanical differences in BGT – such as greater dorsiflexion, reduced ankle plantar flexion, and the reliance on different muscle groups – might lead to enhanced calorie burn and increased metabolic rate.[23] The novel recruitment of posterior leg muscles, in particular, requires concentric contractions that differ from the eccentric contractions during forward gait, potentially offering a more intense workout.[24] Over time, these changes might positively impact body composition markers metabolic markers by engaging different muscle fibers and promoting better muscular utilization of available metabolites.[22,25] Moreover, as emerging research suggests, the effects of gait training on inflammatory markers cannot be overlooked.[26] Inflammatory processes play a crucial role in various chronic diseases and are influenced by physical activity.[27,28] While FGT has shown promise in reducing systemic inflammation,[29] the potential differential effects of BGT on inflammatory markers remain an area of exploration.

While a considerable number of systematic reviews and meta-analyses exist, evaluating the mechanisms and benefits of BGT and FGT, especially in neurological and musculoskeletal rehabilitation,[30,31,32,33,34,35,36] a glaring gap is evident. No existing review rigorously compared BGT to FGT on the body composition, cardiopulmonary fitness, inflammation, and metabolic outcomes. Therefore, this review aims to address the research question: In adults (Population), how does BGT (Intervention) compare to FGT (Comparison) in affecting body composition, cardiopulmonary fitness, inflammatory, and metabolic markers (Outcomes)? By conducting a thorough synthesis of the current literature, this review seeks to fill the existing knowledge gap and highlight potential directions for future research and clinical practice.

METHODS

Registration

The systematic review was conducted as per the recommendations of the Preferred Publishing Items for Systematic Reviews and Meta-Analyses (PRISMA) statement for publishing systematic reviews,[37] to examine the impact of BGT on cardiopulmonary and metabolic markers when compared with FGT.

Search strategy

Five electronic databases – PubMed, Web of Science, PEDro (Physiotherapy Evidence Database), Scopus (Elsevier), and CENTRAL (Cochrane Central Register of Controlled Trials) were searched from the date of inception to November 2023. We identified studies using main search terms by combining key phrases for exposure type (backward walking, backward running, and retro walking) and outcome measurements (body composition, inflammation, metabolic marker, metabolism, heart rate, oxygen uptake, body mass index [BMI], body fat, and waist circumference). In the PubMed and Scopus databases, these search phrases were combined with the Boolean operators OR and AND to broaden or narrow the search results. In the PEDro database, they were coupled with truncation. In addition to searching electronic databases, references to pertinent papers were checked. The complete search strategy for PubMed database was (“backward” OR “retro”) AND (“walking” OR “running” OR “locomotion”) AND (“heart” OR “oxygen” OR “Body composition”). Our search strategy was restricted to English-language publications, focusing on studies with participants aged 18 years or older to align with our target demographic, and including both published and “in press” peer-reviewed articles to capture the most current and relevant findings.

The search was conducted across all databases and obtained articles were imported to the reference manager EndNote™ 21 (Clavirate Analytics), where the results were synthesized, and duplicates were eliminated. The titles of the articles were examined by two authors independently (TA and SZ). Then after, selected articles were screened for inclusion on the basis of their abstracts by another author (SS). Then, full-text analysis was performed for selected articles by another two independent authors (MFA and SMA). Any disagreements were settled through consensus with another reviewer (SN).

Eligibility criteria

To find eligible studies, the populations, interventions, comparators, and outcome framework was used.[38] In the context of this review, we defined our populations as adult humans, interventions as backward gait, comparators as forward gait, and outcomes as indices of body composition, cardiopulmonary, inflammatory, and metabolic markers. Our inclusion criteria encompassed studies with designs such as randomized controlled/clinical trials (RCTs), quasi-experimental trials, cross-over controlled trials, and cross-sectional investigations of the effects of backward gait patterns on body composition, cardiopulmonary, inflammatory, and metabolic markers. The exclusion criteria were research involving individuals below 18 years, studies enrolling participants with musculoskeletal or neurological conditions, and other formats such as review articles, case studies, academic theses, or dissertations and conference proceedings. Data regarding the characteristics of the trial were extracted by two authors independently (ASA and SZ). The authors of the study were contacted if any reported data were not clear or lacked details.

Risk of bias assessment of the methodology

Through domain-based analyses, two authors (AM and SZ) evaluated the quality of RCTs per the Cochrane Collaboration of the risk of bias. Under the following seven domains – random sequence generation, allocation concealment, blinding of participants, personnel and outcome measures, incomplete outcome reporting, selective outcome reporting, and other bias-risk of bias was classified as high, low, or unclear.[39]

Summary of finding table and assessment of the certainty of the evidence

The Grading of Recommendation, Assessment, Development, and Evaluation (GRADE) guidelines were used to assess the methodological quality of the meta-analysis.[40] Two authors (AM and SZ) separately prepared the summary of finding tables for the studies that provided data for the meta-analysis, and any disagreements were settled through discussion with a third author (SS) [Tables 1 and 2]. By evaluating five domains, including (1) risk of bias, (2) inconsistency, (3) indirectness, (4) imprecision, and (5) other considerations, we were able to determine the degree of certainty of the evidence for the pre-specified outcome measures. The GRADEpro GDT software (GRADEpro GDT, McMaster University, and Evidence Prime, 2022) was used for the same.

Table 1.

Backward walking compared to forward walking in adults for body composition outcomes

Patient or population: Adults Setting: Exercise therapy centers Intervention: BW Comparison: FW
Outcomes	Number of participants (studies) follow-up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	Anticipated absolute effects
				Risk with FW	Risk difference with BW
Body density (g/cm3)	60 (2 RCTs)	⨁⨁⨁◯ moderate	-	-	SMD 25.77 lower (77.25 lower–25.71 higher)
Percentage body fat	60 (2 RCTs)	⨁⨁⨁◯ moderate	-	-	SMD 0.11 higher (0.62 lower–0.4 higher)
BMI (kg/m2)	319 (3 RCTs)	⨁⨁⨁⨁ high	-	-	SMD 0.55 higher (0.77 higher–0.32 higher)
Waist-height ratio	213 (2 RCTs)	⨁⨁⨁◯ moderate	-	-	SMD 1.18 higher (1.89 higher–0.48 higher)
Fat mass (kg)	60 (2 RCTs)	⨁⨁⨁◯ moderate	-	-	SMD 0.03 higher (0.47 lower–0.54 higher)
Fat free mass (kg)	60 (2 RCTs)	⨁⨁⨁◯ moderate	-	-	SMD 0.25 lower (0.75 lower–0.26 higher)
Fat free mass (kg)	60 (2 RCTs)	⨁⨁⨁◯ moderate	-	-	SMD 0.25 lower (0.75 lower–0.26 higher)
WHR	213 (2 RCTs)	⨁⨁⨁⨁ high	-	-	SMD 5.74 higher (14.77 lower–4.28 higher)

GRADE working group grades of evidence: High certainty - we are very confident that the true effect lies close to that of the estimate of the effect, moderate certainty - we are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different, low certainty - our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect, very low certainty - we have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect. CI: Confidence interval, SMD: Standardized mean difference, FW: Forward walking, BW: Backward walking, BMI: Body mass index, WHR: Waist–hip ratio, RCTs: Randomized clinical trials

Table 2.

Backward walking compared to forward walking in adults

Patient or population: Adults Setting: Exercise therapy centers Intervention: BW Comparison: FW
Outcomes	Number of participants (studies) follow-up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	Anticipated absolute effects
				Risk with FW	Risk difference with BW
CRP (mg/L)	208 (2 RCTs)	⨁⨁⨁⨁ high	-	-	SMD 0.99 higher (1.28 higher–0.70 higher)

GRADE working group grades of evidence: High certainty - we are very confident that the true effect lies close to that of the estimate of the effect, moderate certainty - we are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different, low certainty - Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect, very low certainty - we have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect. CI: Confidence interval, SMD: Standardized mean difference, FW: Forward walking, BW: Backward walking, RCTs: Randomized clinical trials, CRP: C-reactive protein

Strategies for data synthesis

The appropriate data for the intended outcome measures, including the number of subjects (n), mean and standard deviation, were extracted from each study group at the beginning and end of the intervention. The retrieved data were imported to the Review Manager (Revman 5, The Cochrane Collaboration) to perform the analysis. To compare the effect of backward gait against forward gait on various outcomes, pooling was used to derive the standardized mean difference (SMD), which was subsequently presented as Cohen’s d effect size (ES), 95% confidence interval (CI), and standard error (SE). A random-effects model was employed to compute the pooled SMDs, integrating individual weighted SMDs with their corresponding SEs. This approach was selected due to the inherent heterogeneity anticipated across the included studies, stemming from variations in study populations, intervention specifics, and outcome measures. The formula used for this calculation was SMD = MD/SD pooled, where MD was the mean difference between the backward and forward groups and SD was the standard deviation. To derive the mean differences, the same units of measurement were used for all outcome measures. The magnitude of ES was classified as minor, moderate, or large depending on whether the value was < 0.2, 0.2–0.5, or > 0.5, respectively.[41] The statistical significance was set at 5% (P = 0.05). The Cochrane’s Q-test was used to obtain I² that quantified the heterogeneity of RCTs. If the value of I² was 25%, 25%–75%, and 75%, it was regarded as low, moderate, and high observed variability between estimated effects, respectively.[39] This index was instrumental in guiding our decision to utilize the random-effects model, as it aptly accommodates the variability across studies by assuming that the observed effects are drawn from a distribution of effects rather than a single fixed effect.

RESULTS

Search results

In the process of identifying relevant literature, a total of 6341 records were initially identified through the database search. During the screening phase, duplicates were removed, leaving 1186 records. These records were then screened by their titles, which resulted in the exclusion of 1073 records. From the remaining 113 records, a full-text assessment for eligibility was conducted. After a thorough evaluation, 108 of these articles were excluded due to reasons such as the lack of eligible outcome measures and the absence of a forward-walking group as a comparator. Consequently, a total of 5 studies were deemed suitable and were included in both the qualitative and quantitative synthesis [Figure 1].

Figure 1.

Preferred publishing Items for Systematic Reviews and Meta-Analyses flow diagram, for eligible studies

Study design

Five eligible randomized clinical trials were retrieved. The included studies were published from the year 2014–2023. All of them were conducted in Saudi Arabia (KSA).[42,43,44,45]

Participants

Five of the included trials consisted of 379 participants (all males) with sample sizes ranging from 30 to 111. A common limitation found in all studies was the lack of female participants. All subjects ranged from 18 to 25 years of age. Only three included studies provided information on sample size calculation.[42,44,45] Two studies included participants with normal BMI.[43] One study had subjects with waist-to-hip ratio (WHR) >1,[45] while the other two studies recruited participants in the overweight and/or obese range of BMI[42,44] [Table 3].

Table 3.

Characteristics of the studies included in the review

Study	Participants description	Intervention	Outcome measured	Results
[43]	30 healthy (mean age of 20.93±2.54 years, mean weight of 59.43±5.38 kg, mean height of 1.69±0.05 m)	FWG - 6 min of forward treadmill walking at an inclination of 10° constant speed of 4 mph, 3 sessions per week, for 6 weeks with 1-day rest between sessions BWG - backward treadmill walking at the same speed, inclination, and training schedule	Average power using RAST, body density, percentage body fat, fat mass, fat-free body mass	↑ anaerobic performance (BWG > FWG)
[46]	30 subjects (mean age of 20.93±2.54 years; weight of 59.43±5.38 kg and height of 1.69±0.05 m	FLG - forward locomotion training on a treadmill at 10° inclinations for 6 weeks, 3 days/week with 1 day of rest between training sessions RLG - retro treadmill training at the same inclination, speed, and training schedule	VO2 max, body density, % body fat, fat mass, fat-free body mass	↑ VO2 max (RLG > FLG)
[45]	111 participants with mean age of 21.58±2.17, mean height of 172.17±3.73, mean weight of 81.60±10.10, and mean BMI of 27.54±3.105	FWG - treadmill training with speed 4 km/h, 10% gradient, 15–30 min of training, 5 min of warm up and cool down, 4 times/week for 6 weeks RWG - retro walking treadmill training with similar parameters	BMI, WC, WHtR, HC, WHR	↓ WHR (RWG > FWG) ↓ WHtR (RWG > FWG)
[42]	102 participants with mean age of 21.39±1.54 in RWG and 21.28±1.69 in FWG. Mean BMI in RWG was 33.92±2.80 and FWG was 33.72±3.76	RWG - retro walking treadmill training with speed 4 km/h, 10% gradient, 15–30 min of training, 5 min of warm up and cool down, 4 times/week for 6 weeks. Then 30 min of training for another 6 weeks FWG - forward walking treadmill training with similar parameters	Adiponectin, CRP, BMI, WHtR, and WHR	↑ adiponectin (RWG > FWG) ↓ CRP (RWG > FWG) ↓ BMI (RWG > FWG) ↓ WHtR↓WHR (RWG > FWG)
[44]	106 participants FWG - n=53, mean age - 21.32±1.73, mean BMI - 32.70±4.88 RWG - n=53, mean age - 21.47±1.54, mean BMI - 32.12±4.37	RWG - retro walking treadmill training with speed 4 km/h, 10% gradient, 15–30 min of training, 5 min of warm up and cool down, 4 times/week for 12 weeks FWG - forward walking treadmill training with similar parameters	BMI, DBP, SBP, CRP	↓ BMI, ↓ DBP, ↓ SBP, ↓CRP RWG > FWG for all

FWG: Forward walking group, BWG: Backward walking group, FLG: Forward locomotion group, BLG: Backward locomotion group, RWG: Retro walking group, RAST: Running-based anaerobic sprint test, VO₂ max: Maximum oxygen uptake, BMI: Body mass index, WC: Waist circumference, HC: Hip circumference, WHR: Waist-hip ratio, WhtR: Waist–height ratio, DBP: Diastolic blood pressure, CRP: C-reactive protein, SBP: Systolic blood pressure, ↑: Increase in value, ↓: Decrease in value

Interventions

All studies investigated and compared the effects of FGT and BGT. The duration of training ranged from 6 weeks.[43,44,45] Each exercise session varied from 7 min to 30 min. Two studies set their frequency at 3 sessions/week.[43,46] While in three studies participants performed training for 4 days/week.[42,44,45] Treadmill speed was set constant at 4 miles/h with 10% inclination in two studies.[43,46] The remaining studies had treadmill speed set at 4 km/h with 10% inclination.[42,44,45] The participants have performed FGT and BGT on treadmill in the included studies.[42,43,44,45,46] All studies used precautionary measures such as familiarizing sessions and supervised training.

Outcome measures

One study focused on anaerobic performance assessed using a sprint test,[43] while another included study measured the aerobic capacity of participants using the volume of maximum oxygen consumption (VO2 max).[46] In addition, these studies also analyzed percentage body density, body fat (%BF), fat mass (FM), and fat-free mass (FFM) as body composition markers.[43,46] Two included studies examined BMI and WHR values to assess body composition.[44,45] Two studies have included the evaluation of waist-to-height ratio (WhtR) and C-reactive.[42,44]

Risk of bias assessment

Figures 2 and 3 illustrate a summary of the risk of bias for each of the domains listed below.

Figure 2.

Risk of bias summary: Review authors’ judgements about each risk of bias item for each included study

Figure 3.

Risk of bias graph: Review authors’ judgements about each risk of bias item presented as percentages across all included studies

Selection bias

Random sequence generation

All the included trials have mentioned randomization. However, only 3 studies reported a method of randomization using sequentially numbered envelopes,[42,44,45] and were thereby considered as low risk of bias. Two studies did not report any method of randomization so were considered at unclear risk of bias.[43,46]

Allocation concealment

Three studies reported a method of allocation using opaque sealed envelopes,[42,44,45] and were put under low risk of bias. The remaining two studies did not provide a method of allocation concealment,[43,46] and were considered under the category of unclear risk of bias.

Performance bias

Blinding of participants and personnel

None of the studies mentioned any information related to the blinding of participants and personnel,[42,43,44,45,46] hence all of them were considered as unclear risk of bias.

Detection bias

Blinding of outcome assessment

Only two studies reported that a separate assessor trained by the authors evaluated the outcome measures and therefore these studies were put under low risk of bias.[42,44] The remaining three studies did not provide any information on the blinding of the assessor and were considered an unclear detection bias.[43,45,46]

Attrition bias

Incomplete outcome data

Two studies did not report any information about drop-outs; hence, they were considered under unclear attrition bias.[43,46] Three studies mentioned number of dropouts and analyses was as per the intention-to-treat principle and, therefore were considered a low risk of bias.[42,44,45]

Selective reporting

Of the five included studies, only three were recorded on the clinical trials registry.[42,44,45] However, all trials reported outcome measure results at all time points and were considered as low risk of bias.[42,43,44,45,46]

Other sources of bias

Other potential risk of bias was adopting a convenient sample of participants[42,43,44,45,46] and these studies were considered at high risk of bias.

Effect of backward gait training

Kachanathu et al. reported a greater positive change in the anaerobic performance of the backward walking group after 6 weeks of treadmill training.[43] A similar improvement was found in aerobic performance in a study by another included study. Both studies did not observe any significant difference in body composition measures.[43,46] Three studies reported larger improvements in backward walking groups for BMI values,[42,44,45] while two studies demonstrated significant improvement in WHR and WhtR following BGT.[42,44,45] Moreover, three studies reported an increase in adiponectin levels after 6 weeks of backward walking training[42,44,45] and found a greater decrease in C-reactive protein levels after 6–12 weeks of BGT.

Pooled results

The following outcome measures of body composition and inflammatory markers were included in the meta-analysis: body density, %BF, FM, FFM, BMI, WhtR, WHR, and C-reactive protein [Figure 4 and Table 1].

Figure 4.

Results of the meta-analysis and forest plots for body composition: (a) Forest plot of comparison: forward walking, outcome: Body density (g/cm3), (b) Forest plot of comparison: forward walking, outcome: Percentage body fat (%), (c) Forest plot of comparison: forward walking, outcome: Fat mass (kgs), (d) Forest plot of comparison: forward walking, outcome: Fat free mass (kgs), (e) Forest plot of comparison: forward walking, outcome: BMI (kg/m2), (f) Forest plot of comparison: forward walking, outcome: Waist – height ratio, (g) Forest plot of comparison: forward walking, outcome: Waist – hip ratio

The pooled analysis demonstrated an insignificant large reduction in body density (SMD-25.77, 95% CI-77.25–25.71, I² = 98%, P = 0.33). A minor insignificant decrease was observed in values of %BF (SMD-0.11, 95% CI-0.62–0.40, I² = 0%, P = 0.67), minor insignificant reduction in FM (SMD-0.03, 95% CI-0.047–0.54, I² = 0%, P = 0.90), and moderate insignificant reduction in FFM (SMD-0.25, 95% CI-0.75–0.20, I² = 0%, P = 0.34). Body density revealed high heterogeneity while %BF, FM, FFM had low heterogeneity between studies. A moderate significant decrease was observed in BMI (SMD-0.55, 95% CI-0.77–0.32, I² = 0%, P < 0.01), and a large significant decrease in WhtR (SMD-1.18, 95% CI-1.89–0.48, I² = 83%, P < 0.01). The WHR revealed large insignificant decrease (SMD-5.74, 95% CI-15.77–4.28, I² = 99%, P = 0.26). Heterogeneity between trials was high in WHR and WhtR and low in BMI. Finally, C-reactive protein demonstrated a significant large decrease with low heterogeneity (SMD-0.98, 95% CI-1.28–0.70, I² = 0%, P < 0.01) [Figures 4 and 5].

Figure 5.

Results of the meta-analysis and forest plots for inflammatory markers: (a) Forest plot of comparison: forward walking, outcome: C – reactive protein (mg/L)

Sufficient data was not available for cardiopulmonary fitness. Thus, the pooled analyses were performed only for body composition and inflammatory markers.

DISCUSSION

The present systematic review and meta-analysis synthesized available evidence comparing the effects of BGT to FGT on various health markers in adults. The findings of the meta-analysis demonstrated statistically significant changes in body composition markers, including WhtR and BMI, and in the inflammatory marker, C-reactive protein levels were reduced significantly. The qualitative analyses revealed improvements in cardiopulmonary fitness, body composition, and inflammatory and metabolic markers.

Body composition

Studies included in the meta-analysis provided mixed findings regarding body composition. A crucial observation is the difference in study durations. Soman et al. demonstrated no changes in BMI after 6 weeks but reported a significant decrease after 12 weeks.[45] Similarly, a pronounced effect on body composition was noted in obese participants after 6 weeks.[42] However, Kachanathu et al. and Kachanathu et al. concluded that backward walking does not change body composition parameters after 6 weeks of training.[43,45,46] The possible mechanism for these findings might be attributed to the already healthy BMI range of participants and the relatively short intervention period.[43,46] Further, the findings of meta-analysis revealed a significant decrease in WhtR. The WhtR is a valuable anthropometric measure that has gained attention as a predictor of metabolic health and cardiovascular risk.[47,48] The improvement in WhtR observed in the present review suggests that BGT may have a positive impact on body composition and metabolic health. The reduction in waist circumference relative to height can signify a decrease in central adiposity, indicating a potential reduction in the fat stored around internal organs, referred to as the visceral fat.[47,49] Visceral fat is known to be metabolically active and is associated with increased inflammation, insulin resistance, and adverse lipid profiles.[50,51]

The underlying mechanism for better improvement in the BGT group on body composition outcomes may be due to the modification in neuromuscular coordination.[52] Forward gait is a habitual movement pattern, and might not pose the same level of neuromuscular challenge as backward gait, however, the neural controls for forward walking are well-established and efficient.[12,53] Previous studies have suggested novel activities may necessitate increased motor unit recruitments, elevating energy demands.[54] This is supported by ample evidence indicating higher physiological demands for backward locomotion compared to forward locomotion under similar conditions.[25,55,56,57] The higher physiological demand manifest as enhanced calorie expenditure, augmented muscle fiber activation, and a more robust mobilization of fat reserves.[58] Over time, as the body continuously engages with these heightened demands, there will be significant improvement in body composition.[59,60] Hence, the BGT’s inherent ability to impose heightened physiological challenges might offer a promising avenue for interventions aiming at improved body composition.

Cardiopulmonary fitness

The results of this review revealed heterogeneous findings regarding the impact of BGT on cardiopulmonary fitness. Kachanathu et al. reported a significant improvement in anaerobic performance following 6 weeks of BGT, whereas Kachanathu et al. demonstrated an increase in aerobic capacity measured by VO₂ max in response to BGT.[43,45,46] These findings align with the notion that BGT can induce adaptations in cardiorespiratory fitness. Previous evidence supports the idea that walking backward engages different muscle groups and movement patterns compared to forward walking, potentially leading to cardiovascular adaptations.[21,61]

The underlying mechanism for the observed improvements in cardiopulmonary fitness could be attributed to the novelty and increased demand placed on muscles during backward walking.[61] The altered muscle recruitment patterns and increased metabolic demand may lead to enhanced cardiovascular responses, such as increased heart rate and improved oxygen uptake.[55] Moreover, the engagement of additional stabilizing muscles during backward walking might contribute to improved overall cardiovascular fitness.[62] However, these findings should be interpreted cautiously due to the limited number of studies and potential bias introduced by small sample sizes and lack of female participants. Further research with larger and more diverse populations is warranted to establish the consistent effects of BGT on cardiopulmonary fitness.

Metabolic and inflammatory markers

The studies included in this review provided valuable insights into the effects of BGT on inflammatory and metabolic markers, shedding light on potential avenues for enhancing metabolic health. Notably, the studies revealed mixed results, similar to the findings in body composition and cardiopulmonary outcomes. Alkhathami et al., Soman et al. and Soman et al. reported improvements in adiponectin levels and a decrease in C-reactive protein levels following BGT, suggesting potential positive metabolic adaptations.[42,44,45] These changes are particularly relevant as adiponectin is associated with improved insulin sensitivity and reduced inflammation, while elevated C-reactive protein levels are linked to metabolic dysfunction.[63,64] The underlying mechanism for the observed improvements in metabolic markers could be attributed to the distinct neuromuscular necessities of BGT.[52] The higher motor unit recruitment and metabolic demand may lead to the mobilization of fat reserves.[54,58] This may contribute to improvements in insulin sensitivity and reductions in systemic inflammation, subsequently influencing adiponectin and C-reactive protein levels.[63] However, it is important to note that while C-reactive protein levels exhibited a decline in this meta-analysis, there was insufficient data to comprehensively assess the effects of BGT on other metabolic markers. Further research with larger sample sizes and comprehensive metabolic assessments is needed to provide a clearer understanding of these effects.

Strengths

The strengths of this study lie in its comprehensive approach to synthesizing existing literature and assessing the effects of BGT across multiple health markers. The rigorous search strategy, adherence to PRISMA guidelines, and assessment of risk of bias enhance the validity and reliability of the findings. The inclusion of a meta-analysis further strengthens the evidence by providing a pooled estimate of the effects, despite the heterogeneity observed among the studies.

Limitations

However, this study is not without limitations. The limited number of eligible studies, predominantly conducted in a single geographical location and involving predominantly male participants, limits the generalizability of the findings. The variations in study design, intervention protocols, and outcome measures also introduce heterogeneity, which might influence the overall interpretation of the results. Another limitation was the absence of blinding in some studies. Furthermore, the publication bias was not evaluated as the sample size for meta-analysis was small, additionally, studies are warranted to affirm our findings.

Future recommendations

BGT can be used as a modification of FGT in various aerobic training interventions to improve cardiopulmonary fitness. It can be added to obesity management programs for weight reduction and disease prevention. There is a need for more high-quality controlled trials based on backward gait patterns. Additional studies are required on a larger sample size for developing protocol and dose-response relationship of BGT in various population. In essence, while this study highlights the potential advantages of BGT, it also emphasizes the need for more rigorous and diverse research to ascertain its true effectiveness and utility in promoting holistic well-being in the adult population.

CONCLUSION

The findings suggest that BGT may offer advantages, particularly in terms of body composition improvements and potential enhancements in cardiopulmonary fitness, and inflammatory and metabolic markers. The BGT could be a promising alternative or complementary approach to conventional FGT for promoting overall health and well-being.

Research quality and ethics statement

This project did not require approval by the Institutional Review Board/Ethics Committee. The project was prospectively registered with PROSPERO (CRD42023398199; registration date: February 9, 2023). The authors adhered to the applicable EQUATOR Network guidelines, specifically the PRISMA 2020 statement, during the conduct of this research project.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.