Effects of resistance training on muscle strength, body composition and immune-inflammatory markers in people living with HIV: a systematic review and Meta-analysis of randomized controlled trials

Abstract

Purpose: The purpose of this review is to evaluate the effect of resistance training (RT) as a unique intervention on muscle strength, body composition, and immune-inflammatory markers in people living with HIV (PLHIV).

Methods: The searches were conducted in seven databases and included published randomized clinical trials that assessed the effect of RT vs. no exercise on muscle strength, body composition, and immune-inflammatory markers in PLHIV until June 2021. Random effects meta-analyses of mean differences (MD) and their 95% confidence intervals (CI) were performed, and the effect size was estimated by Hedges’ g test.

Results: Seven RCTs were included (n= 258 PLHIV) and the study duration lasted between six and 24 weeks. In comparison to no exercise, RT improved muscle strength in bench press (MD 10.69 kg, 95%IC 3.44 to 17.93, p= 0.004, g =2.42) and squat (MD 22.66 kg, 95%IC 7.82 to 37.50, p= 0.003, g = 3.8) exercises, lean body mass (MD 2.96 kg, 95%CI 0.98 to 4.94, p= 0.003, g = 1.99), fat body mass(MD −2.67 kg; 95%CI −4.95 to −0.39, p= 0.02, g=-0.99), body fat percentage (MD −3.66%, 95%CI −6.04 to −1.29, p= 0.003, g=-1.99) and CD4⁺ cells count(MD 100.15 cells/mm³, 95%CI 12.21 to 188.08, p = 0.03, g = 2.91) in PLHIV. There was no effect of RT on IL-6 (MD −1.18 pg/mL, 95%CI −3.71 to 1.35, p = 0.36, g = 0.001) and TNF-α (MD −4.76 pg/mL, 95%CI −10.81 to 1.29, p = 0.12, g=-1.3) concentrations in PLHIV. Conclusions: RT as a unique intervention improves muscle strength, body composition and CD4⁺ count cells in PLHIV.

Keywords:

• HIV/AIDS
• strength training
• lean body mass
• lymphocyte CD4⁺
• TNF-α
• IL-6

Introduction

People living with HIV (PLHIV) have significantly increased life expectancy after the highly antiretroviral therapy (HAART) era and nowadays the infection is considered a chronic disease.¹ However, HIV infection and HAART use are related to a large number of immune-metabolic and morphologic alterations such as chronic inflammation, dyslipidemia, diabetes, cardiovascular diseases, lipodystrophy, and wasting muscle.^2–7 Furthermore, has been reported that PLHIV presents lower physical capacities (e.g. muscle strength and/or cardiorespiratory fitness) compared to the general population which contributes to poor aging and quality of life.⁸ Thereby, interventions to reduce and/or prevent the side effects of infection and HAART use became necessary for PLHIV.

In this scenario, exercise training plays a key role in the chronic management of PLHIV. Specifically, resistance training (RT) is characterized by using a resistance apparatus such as gym equipment, free weights, and elastic tapes against the target muscle group contraction.⁹ Furthermore, there is a consensus that RT programs enhance muscle mass, muscle strength, and improves immune-metabolic markers in the general population.⁹

Although there was an increase in studies involving PLHIV and exercise training, the results obtained by RT programs as unique intervention remain inconsistent.^10–16 The previous meta-analysis evaluated the effects of RT in PLHIV however were included results obtained by studies with combined exercise training (aerobic and resistance training in the same session) which could be a confusion factor for outcomes.^17–21 Although these previous meta-analyses found benefits for PLHIV it is important to evaluate the effects of RT program as a unique intervention for this population. This systematic review and meta-analysis aimed to determine the effects of RT on muscle strength, body composition, and immune-inflammatory markers in PLHIV when compared with no exercise in randomized controlled trials (RCTs).

Methods

This systematic review and meta-analysis were conducted following the recommendations outlined in Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) statements²² and registered in the PROSPERO database (CRD42020182343).

Eligibility criteria

Articles were eligible for inclusion if they were RCTs involving PLHIV (18-60 years old) undergoing HAART (for at least 2 years) in which RT was carried out as a unique exercise intervention, performed two times a week and for four weeks at least, compared with no exercise (control group) and where outcomes were: (1) muscle strength (assessed via one-repetition maximum) in the bench press and/or squat exercise, (2) body composition measured by lean body mass (LBM), fat body mass (FBM) and/or body fat percentage (BF%) (assessed via dual-energy X-ray absorptiometry, bioelectrical bioimpedance or skinfold thickness), lymphocyte TCD4⁺ cell counts (via flow cytometric), interleukin-6 (IL-6) and tumor necrosis factor-alpha (TNF-α) (via enzyme-linked immunosorbent assay or cytometric bead array). Studies were ineligible if involving anabolic steroids or nutritional supplementation, reviews, and systematic reviews, and those that investigated the RT acute effects.

Search strategy

Searches were performed in PubMed, Europe PMC, Web of Science, Scielo, SPORTDiscuss, PEDro and Cochrane databases without language restriction until June 2021. A Medical Subject Heading (MeSH) search was performed using the keywords ‘resistance exercise’ and ‘HIV’. Thus, the following search strategy was used to search articles: (‘Resistance training’ OR ‘Strength Training’ OR ‘Weight-Lifting Strengthening Program’ OR ‘Weight-Lifting Strengthening exercise’ OR ‘Weight-Bearing Strengthening Program’ OR ‘Weight-Bearing Strengthening exercise’) AND (‘HIV’ OR ‘people living with HIV’ OR ‘seropositive people’).

Studies selection

In the initial screening, two independent authors (HRZ and LTPL) performed the search and screened the titles, abstracts, or full texts from database records. The duplicates and the studies that did not meet the eligibility criteria were removed. In the case of disagreements between the authors a third author was included (AG).

Data extraction

Two researchers (VLS and WFS) independently extracted data using a customized spreadsheet. After the data extraction, there was a conference of the data, and a third researcher (ELM)was consulted in case of disagreement. The data extracted from the article were (1) number of initial participants involved in the study; (2) number of participants that finished the intervention; (3) training protocol, including week frequency, number of exercises, volume, and intensity; (4) outcomes per arm. Quantitative data from pre- and post-training intervention were extracted from text, tables, and figures of published articles as mean ± standard deviation (SD), and in instances where results were presented as the standard error of the mean (SEM), it was converted to SD.

Quality assessment

Two researchers (VLS and WFS) independently assessed the quality of the included studies using the risk of bias 2.²³ The tool consists of five distinct domains and leads to judgments of “low risk of bias”, “high risk of bias” and/or “some concerns”.

Rating quality of evidence

The Grading of Recommendations Assessment Development and Evaluation (GRADE), which evaluated the risk of bias, consistency; directness; precision; and publication bias, was used to determine the quality of evidence per outcome.²⁴ The quality of evidence was graded from ‘high quality’ to ‘very low quality’ according to the criteria.

Statistical analysis

Meta-analyses were performed using Review Manager 5.3 for Windows. The delta (post-pre) values were calculated for all outcomes and the SD change was calculated according to the equation $SD\mathrm{ }\mathit{change}=\surd \left[\right(\mathit{SDpre})2+(\mathit{SDpost})2-(2\mathrm{ }x\mathrm{ }\mathit{corr}\mathrm{ }x\mathrm{ }\mathit{SDpre}\mathrm{ }x\mathrm{ }\mathit{SDpost}\left)\right]$ where the imputed correlation coefficient is 0.50.²⁵ Effects on outcomes are presented as mean differences (MD) and their 95% confidence intervals (CI). Study heterogeneity was assessed using Q and I² statistics. The heterogeneity thresholds for I² were 25% (low), I² = 50% (moderate), and I² = 75% (high) .²⁶ Furthermore, the mean effect size was estimated by Hedges’ g test. Due to anticipated study heterogeneity, random-effects models were used. Publication bias was examined visually via funnel plots and statistically (p < 0.10) using Egger’s.

Results

Search results

The initial database search found 106 articles and 2 articles were included from other sources. A total of 57 articles were removed due to duplicate records and two articles were excluded from records screened. Forty-seven studies were assessed for eligibility which 40 were excluded based on (a) studies involving aerobic exercise in the training protocol (n= 29); (b) intervention with PLHIV under 18 years (n= 1); (c) study involving anabolic steroid or nutritional supplementation (n= 7); (e) not evaluated the outcomes target of the present study (n= 3). Finally, seven RCTs (n= 258) were included in the meta-analysis.^10–16 The PRISMA flowchart is demonstrated in Figure 1.

Figure 1. PRISMA flow diagram and selection of articles.

Descriptive characteristics of the studies

Included RCTs are summarized in Table 1. Four studies were conducted in Brazil,^10,12–14 one in the United States of America,¹⁵ one in Iran¹¹ and one in Zimbabwe.¹⁶ Two RCTs involved the same sample^12,14 and therefore a total of 261 PLHIV were recruited to RT programs and 258 (95 men and 163women) completed follow-up, representing a loss of 1.1%. Five studies reported the mean age of participants (41.04 ± 0.82 years).^11–15 Four studies reported the time of infection diagnosis (8.5 ± 2.2 years) .^11–14 All studies included participants under HAART use and presented undetectable viral load.

Table 1. Summary of included studies.

Study	Initial Study Design	Final participants	Male /Female	Age years (M ± SD)	Training Protocol	Outcomes Results*
Brito 2013 ^10	45 PLHIV randomized to RT or CON	RT n = 23; CON n = 22	14/31	NR	24 weeks of RT; 3x/week; 7 exercises; 3 sets; 8–10 reps; 80% 1RM	↑ Muscle strength ↔ LBM ↓ FBM ↓ BF% ↔ CD4+
Ghayomzadeh 2019 ^11	21 PLHIV randomized to RT or CON	RT n = 12; CON n = 6	12/6	39 ± 9	8 weeks of RT; 3x/week; Circuit training; phase 1 (week1 to 4): 2 to 3 circuit laps; 8 exercises; 12 to 15 repetitions using elastic band and body weight; phase 2 (week 5 to 8): 3 circuit lap; 10 exercises; 8 to 12 repetitions using elastic band and body weight	↑ LBM ↓ FBM ↓ BF% ↑ CD4+
Vingren 2018 ^15	16 PLHIV randomized to RT or CON	RT n = 8; CON n = 8	16/0	41 ± 11	6 weeks of RT; 3x/week; 13 exercises. 3 to 5 sets; 5 to 12 repetitions with the individual load.	↑ Muscle strength ↔ LBM ↓ FBM ↓ BF% ↔ IL-6 ↔ TNF-α
Zanetti 2016a ^14	30 PLHIV randomized to RT or CON	RT n = 15; CON n = 15	17/13	41.1 ± 10.1	12 weeks of RT; 3x/week; 6 exercises; 1^st session: 3 sets of 4–6 RM; 2^nd session: 3 sets of 15–20 RM; 3^rd session: 3 sets of 8–12 RM	↑ LBM ↓ FBM ↓ BF%
Zanetti 2016 b ^12	30 PLHIV randomized to RT or CON	RT n = 15; CON n = 15	17/13	41.1 ± 10.1	12 weeks of RT; 3x/week; 6 exercises; 1^st session: 3 sets of 4–6 RM; 2^nd session: 3 sets of 15–20 RM; 3^rd session: 3 sets of 8–12 RM	↑ Muscle strength ↑ CD4+ ↓ IL-6 ↓ TNF-α
Zanetti 2017 ^13	21 PLHIV with metabolic syndrome randomized to RT or CON	RT n= 10; CON n = 11	9/12	43 ± 10.5	12 weeks of RT; 3x/week; 6 exercises; 1^st session: 3 sets of 4–6 RM; 2^nd session: 3 sets of 15–20 RM; 3^rd session: 3 sets of 8–12 RM	↑ LBM ↓ FBM ↓ BF%
Sookan 2020 ^16	128 PLHIV randomized to RT or CON	RT = 64; CON = 64	27/101	NR	12 weeks of RT; 3x/week; 4 exercises; 1-2 weeks (3 sets of 10 repetition with 50% 1RM); 3-4 weeks (3 sets of 10 repetition with 60% 1RM); 5-6 weeks (3 sets of 10 repetition with 65% 1RM) ; 7-8 weeks (2 sets of 8 repetition with 70% 1RM); 9-10 weeks (2 sets of 8 repetition with 75% 1RM); 11-12 weeks (2 sets of 8 repetition with 80% 1RM)	↑ Muscle strength ↑ LBM ↓ FBM ↓ BF%

* Compared to the CON group. NR: not reported; LBM: lean body mass; FBM: fat body mass; BF%: body fat percentage; IL-6: interleukin-6; TNF-α: tumor necrosis factor alpha.

Four studies evaluated the muscle strength using the one-repetition maximum test in bench press exercise^10,12,15,16 while three studies also used the squat exercise.^10,12,16 Six RCTs estimate the body composition which five RCTs used skinfolds thickness^10,13–16 and one study used bioelectrical bioimpedance.¹¹ Three studies analyzed the CD4⁺ cells count, and all used the cytometric flow method.^10–12 Finally, two RCTs evaluated the concentrations of IL-6 and TNF-α using cytometric bead array method^12,15

Descriptive characteristics of interventions

Although all RT protocols were performed three times a week, the time of intervention varied between six weeks,¹⁵ eight weeks,¹¹ 12 weeks^12–14,16 and 24 weeks.¹⁰ One study performed circuit training with elastic tape and body weight¹¹ while six studies used machines.^10,12–16 The intervention protocols are described in Table 1.

Quality assessment

The risk of bias graph is demonstrated in Figure 2. All studies presented “some concerns” in the randomization process but overall, the studies presented a low risk of bias.

Figure 2. Risk of bias of included RCTs: (a) risk of bias graph and (b) risk of bias summary. Green circle with plus sign indicates low risk of bias and red circle with minus sign indicates high risk of bias.

Quality of evidence

The GRADE demonstrated a moderate quality of evidence for improvements of muscle strength in the bench press and squat exercise improvements on LBM, FBM and BF%. T CD4⁺ count cells, IL-6 and TNF-α (Table 2).

Table 2. Summary of grading of recommendations assessment development and evaluation (GRADE).

Certainty assessment							Summary of findings
Participants (studies) Follow up	Risk of bias	Inconsistency	Indirectness	Imprecision	Publication bias	Overall certainty of evidence	Study event rates (%)			Anticipated absolute effects
							With control	With RT		Risk with control	Risk difference with RT
Muscle Stregnth in Bench Press
219 (4 RCTs)	serious^a	not serious	not serious	not serious	none	⨁⨁⨁◯ MODERATE	109	110		The mean muscle Stregnth in Bench Press was 0.3 kg	MD 10.69 kg higher (3.44 higher to 17.93 higher)
Muscle Strength for Squat
203 (3 RCTs)	Serious^a	not serious	not serious	not serious	none	⨁⨁⨁◯ MODERATE	101	102		The mean muscle Strength for Squat was 0.3 kg	MD 22.66 kg higher (7.82 higher to 37.5 higher)
Lean Body Mass (LBM)
258 (6 RCTs)	serious^a	not serious	not serious	not serious	none	⨁⨁⨁◯ MODERATE	126	132		The mean LBM was 0.5 kg	MD 2.96 kg higher (0.98 higher to 4.94 higher)
Fat Body Mass (FBM)
242 (5 RCTs)	Serious^a	not serious	not serious	not serious	none	⨁⨁⨁◯ MODERATE	118	124		The mean FBM was 0.39 kg	MD 2.67 kg lower (4.95 lower to 0.39 lower)
Body fat percentage (BF%)
242 (5 RCTs)	Serious^a	not serious	not serious	not serious	none	⨁⨁⨁◯ MODERATE	118	124		The mean BF% was 0.6 %	MD 3.66 % lower (6.04 lower to 1.29 lower)
CD4
93 (3 RCTs)	Serious^a	not serious	not serious	not serious	none	⨁⨁⨁◯ MODERATE	43	50		The mean CD4 was 21.3 cell/mm3	MD 100.15 cell/mm3 higher (12.21 higher to 188.08 higher)
Interleukin-6 (IL-6)
46 (2 RCTs)	serious^a	not serious	not serious	not serious	none	⨁⨁⨁◯ MODERATE	23	23		The mean IL-6 was 0.19 pg/mL	MD 1.18 pg/mL lower (3.71 lower to 1.35 higher)
Tumor Necrosis Factor-alpha (TNF-α)
46 (2 RCTs)	serious^a	not serious	not serious	not serious	none	⨁⨁⨁◯ MODERATE	23	23		The mean tNF-alpha was 0.8 pg/mL	MD 4.76 pg/mL lower (10.81 lower to 1.29 higher)

a (-1) lack of blinding information on study participants.; CI: confidence interval; SMD: standardized mean difference.

Meta-analysis results

Muscle strength

RT program improved muscle strength in bench press (MD 10.69 kg, 95%IC 3.44 to 17.93, p= 0.004, I²= 82%, g= 2.42) (Figure 3a) and squat (MD 22.66 kg, 95%IC 7.82 to 37.50, p= 0.003, I²= 98%, g = 3.8) (Figure 3b).

Figure 3. Effects of RT on muscle strength in PLHIV. (a) bench press exercise; (b) squat exercise.

Body composition

RT program increased LBM (MD 2.96 kg, 95%CI 0.98 to 4.94, p= 0.003, I²= 0%, g = 1.99) (Figure 4a), and reduced FBM (MD −2.67 kg, 95%CI −4.95 to 0.39, p= 0.02, I²= 25%, g= −0.99) (Figure 4b) and BF% (MD −3.66%, 95%CI −6.04 to −1.29, p= 0.003, I²= 28%, g= −1.99] (Figure 4c).

Figure 4. Effects of RT on body composition and imune-inflammatory markers. (a) lean body mass; (b) fat body mass; (c) body fat percentage; (d) T CD4 ⁺ cell counts; (e) interleukin-6; (f) tumor necrosis factor alpha.

Immune-inflammatory markers

RT program improved CD4⁺ cells count (MD 100.15 cells/mm³, 95%CI 12.21 to 188.08, p= 0.03, I²= 0%, g = 2.91) (Figure 4d). However, there were no difference for IL-6 (MD −1.18 pg/dL, 95%IC −3.71 to 1.35, p= 0.36, I²= 93%, g= −0.001) (Figure 4e) and TNF-α (MD −4.76 pg/dL, 95%IC −10.81 to 1.29, p= 0.12, I²= 68%, g= −1.3) (Figure 4f).

Discussion

To the best of our knowledge, this is the first systematic review and meta-analysis that evaluated the effect of RT as a unique intervention for PLHIV. The main results of the present study suggest a beneficial effect of the RT intervention on muscle strength, body composition, and CD4⁺cell count among PLHIV. These results should be interpreted with caution once all studies included clinical stable volunteers with undetectable viral load and HAART use. In addition, the results obtained by this systematic review and meta-analysis were evidenced by RCTs with moderate quality of evidence.

The meta-analysis found beneficial effects of RT on overall and by subgroup muscle strength increase in PLHIV, despite the significant heterogeneity observed among studies. Muscle strength is an important marker associated with the ability to perform daily activities and a mortality predictor.^27,28 It has been reported that PLHIV has a lower muscle strength compared to the non-infected population with up to 57% lower strength in lower limbs.²⁹ In addition, this meta-analysis revealed that RT is an effective intervention for enhancing the body composition in PLHIV and the findings reflect the large benefits such as the lower risk of mortality and higher cardiorespiratory fitness.³⁰ The HIV-related wasting muscle is a well-established phenomenon and is determined by HIV infection, HAART use, and high levels of physical inactivity.^31,32 Previous studies reported that these factors are associated with chronic inflammation and muscle structure abnormalities which are responsible for wasting muscle in PLHIV.^15,29 Metabolic diseases such as insulin resistance and diabetes mellitus 2 are associated with lower LBM therefore RT could be proposed as a key role for prevention or mitigation of these conditions.^33–35

A promising clinical finding of this meta-analysis was the increase in CD4⁺ cell count after RT program in PLHIV. TCD4⁺ lymphocytes play a central role in regulating immune functions by activating B cells in the productions of antibodies and improve cellular immune response to antigens. In this scenario, HIV preferentially infects activated CD4⁺cells resulting in the loss of HIV-specific immune response, recall antibody response and, non-specific immune response in the AIDS stage.³⁶ It has been widely reported the potential of exercise to modulate immune responses in the human being environment. For example, resistance exercise-induced effects on CD4⁺ T activation were observed following an acute RT section and results in a greater susceptibility to HIV‐1 infection in vitro.³⁶ Of course, these results should be observed with caution, as the participants were on HAART use and with a controlled viral load. Furthermore, the studies used for this analysis were classified as of moderate quality, which limits the accuracy of these results.

Our analyses revealed no statistically significant effects of the RT on IL-6 and TNF- levels in PLHIV. The meta-analysis by Ibeneme et al found similar results when including articles that used combined or isolated exercise in PLHIV. It is worth noting that both meta-analyses were conducted with a small number of articles and, thus, conclusions should be taken with caution in this direction.²¹

The present study presents some limitations. First, it is important to focus that RT as a unique intervention for PLHIV is scarce and, therefore, the included studies showed diversified protocols using different times of intervention, volume and intensities which could interfere in the main outcomes. Second, only two studies evaluated inflammatory biomarkers (IL-6 and TNF-α) which could underestimate the anti-inflammatory effect of exercise. Third, most variables evaluated in the present study showed high heterogeneity (I²> 75%) due to the influence of clinical and methodological variability however the problem was solved by the funnel plots.

The present systematic review and meta-analysis clarify that RT should be included in the healthcare programs for PLHIV. Furthermore, the study evidences the scarcity of RCTs involving RT and their variables (e.g., volume, intensity, frequency, periodization, time of intervention) which could be investigated for new studies. Another important point that needs attention is the additional effect of exercise on HAART side effects such as dyslipidemia, diabetes, mitochondrial dysfunction, cardiovascular diseases and mortality.

Conclusion

In conclusion, based on the findings of this meta-analysis, RT may be useful in the management of HIV/HAART-related morphological and immune disorders due to improvement in body composition, muscle strength and CD4⁺ cell count.