3.1Literature retrieval results
Based on the retrieval tactics, a total of 950 studies were searched in this study, including PubMed (n=261), Embase (n=324), Web of science (n=267), and Cochrane Library (n=98). After removing duplicates, 519 studies were initially included, and after excluding literature such as reviews, systematic evaluations, and Meta-analyses, 468 studies were obtained after initial screening, and 59 studies were included after excluding literature with inconsistent study content or interventions by reading titles and abstracts, after which a total of 11 studies that met the criteria were finally included by reading the full text, and systematic review and Meta-analyses were performed The PRISMA flow chart provides details of the reasons for study selection and exclusion of articles, as detailed in Figure 2.
The reasons for exclusion of 52 studies contained:
1. the indicator was EMG of hamstring muscle activity (n=13), a systematic review of the effect of NHE on hamstring muscle activity has been conducted in a previous study [19].
2. the index is the incidence of HSI (n=5), the previous study [18] has conducted a Meta-analysis on the incidence of hamstring injury in soccer players by NHE.
3. the index is human balance function (n=4)
4. the indicator is the change of red blood cells (n=1)
5. the index was kinematic analysis (n=2)
6. incomplete data(n=19)
7. full text not available(n=4).
3.2 Basic characteristics of included studies
The study involved 497 participants, ranging in age from 19 to 30 years old. The basic information, sample size of subjects, intervention programs, clinical indicators, etc. from the included studies were extracted. In the 11 included studies[22][23][24][25][26][27][28][29][30][31][32], the intervention methods in the experimental group were all NHE, and the control group was conventional training or training placebo. All studies confirmed that the subjects were in a healthy state and had no sports injuries, and the intervention period ranged from 4-27 weeks. See table 3 for details.
Table 3: Basic information of included research literature
Study
|
Location
|
Population
|
Subjects and gender(M/F)
|
Subject's age
|
Duration (weeks)
|
Intervention (reps × sets × frequency)
|
Outcomes
|
test group
|
control group
|
test group
|
control group
|
test group
|
control group
|
Iga et al. [22]
|
Britain
|
professional soccer players
|
10M
|
8M
|
23.4±3.3
|
22.3±3.9
|
4
|
1-3×2-3×5-12
|
Φ
|
C,M
|
Delextrat et al. [23]
|
Britain
|
university-level hockey players
|
9F
|
8F
|
19.7±1.4
|
19.6±1.4
|
6
|
3×2-3×6-10
|
Φ
|
C,H,M
|
Suarez-Arrones et al. [24]
|
Spain
|
professional football players
|
group 1:16M
|
17M
|
18.8±0.8
|
17
|
1-2×2-3×5-10
|
Φ
|
E,F
|
group 2:17M
|
15
|
Ishøi et al. [25]
|
Denmark
|
amateur football players
|
18M
|
17M
|
---
|
10
|
1-3×2-3×5-12
|
Φ
|
C,E
|
Ribeiro-Alvares et al. [26]
|
Brazil
|
healthy young adults
|
3M/7F
|
3M/7F
|
23.7 ± 3.3
|
26.0 ± 2.7
|
4
|
2×3×6-10
|
Φ
|
A,C,G,H,I,J,K,N
|
Seymore et al. [27]
|
America
|
healthy young adults
|
4M/6F
|
2M/8F
|
18.3 ± 0.5
|
19.9 ± 1.2
|
6
|
1-3×2-3×5-12
|
Φ
|
B,D,J,K,L,M,O
|
Delahunt et al. [28]
|
Republic of Ireland
|
healthy young adults
|
15M
|
14M
|
22±1.38
|
6
|
1-3×2-3×5-12
|
Φ
|
C,D,M
|
Vianna et al. [29]
|
Brazil
|
professional soccer players
|
17F
|
16F
|
24±5
|
24±4
|
8
|
2×2-4×6-10
|
Φ
|
E,K
|
Salci et al. [30]
|
Turkey
|
recreational athletes
|
13F
|
12F
|
20.5±4.9
|
21.0±1.6
|
10
|
1-3×2-3×5-12
|
Φ
|
B,D
|
Siddle et al. [31]
|
United Kingdom
|
amateur Soccer and Rugby players
|
7M
|
7M
|
20.13±1.55
|
20.86±1.57
|
6
|
1-2×2-3×5-10
|
Φ
|
C
|
Hasebe et al. [32]
|
Japan
|
high school soccer club level players
|
156M
|
103M
|
16.7 ± 0.5
|
16.3 ± 0.6
|
27
|
1-2×2-3×6-10
|
Φ
|
G
|
“---”: Not stated; reps: repetitions; “Φ”: normal training or placebo; Outcomes: “A”: The concentric hamstring peak torque (conH PT); “B”: The normalized concentric hamstring peak torque (Normalized conH PT); “C”: The eccentric hamstring peak torque (eccH PT); “D”: The normalized eccentric hamstring peak torque (Normalized eccH PT);“E”: The eccentric hamstring strength peak; “F”: The normalized eccentric hamstring strength peak; “G”: The conventional H:Q ratio; “H”: The functional H:Q ratio ;“I”: BFLH thickness; “J”: BFLH Pennation angle; “K”: BFLH Fascicle length; “L”: BFLH volume; “M”: The hamstrings eccentric angle of peak torque(APT);“N”: The hamstring flexibility ;“O”: The hamstring stiffness.
3.3 Results of the quality evaluation of the included studies
The quality evaluation of the included 11 articles was performed using Review manager 5.3 software, and the results are shown in Figures 3 and 4. The evaluation entries included selection bias, performance bias, detection bias, attrition bias, reporting bias, and other bias. bias, reporting bias, and other bias, with selection bias subdivided into random sequence generation and allocation concealment. All included literature had a total score of ≥4, and the overall quality was high.
3.4 Meta-analysis results
In this study, the clinical indicators of each experiment were summarized and found to be 15 indicators, as shown in Table 4. It was found that there was only one study reported on six indicators, including the concentric peak torque of the hamstrings[26]; the eccentric hamstring strength peak to body weight[33]; BFLH muscle thickness[26];BFLH volume[27]; flexibility of hamstrings[26]and hamstrings muscle stiffness[27], and three other indicators were reported in two studies,such as the concentric peak torque of the hamstrings muscles to body weight[27, 28],H:Q conventional ratio[26, 32],BFLH Pennation angle[26, 27].But the heterogeneity was large, and nine indicators such as the overview could not be Meta-analyzed.
Therefore, in this paper, Meta-analysis was performed on six clinical indicators including the eccentric hamstring strength peak, the eccentric peak torque of the hamstrings,the eccentric peak torque of the hamstrings muscles to body weight, the hamstrings eccentric angle of peak torque(APT),fascicle length of biceps femoris, and H:Q functional ratio in the included studies. Since some of the studies tested the peak moments of different groups or hamstrings in low [≤60 (°)/s], medium [120 (°)/s], and high [≥180 (°)/s] angular velocity conditions, the data from different trials of the same study were expressed as T1, T2···Tn.
Table 4:Statistical details of the clinical indicators for each experiment
Statistics of clinical indicators of each experiment(n=15)
|
- the concentric peak torque of the hamstrings
- the concentric peak torque of the hamstrings muscles to body weight
- the eccentric peak torque of the hamstrings
- the eccentric peak torque of the hamstrings muscles to body weight
- the hamstrings eccentric angle of peak torque(APT)
- the eccentric hamstring strength peak
- the eccentric hamstring strength peak to body weight
- H:Q conventional ratio
- H:Q functional ratio
- BFLH thickness
- BFLH pennation angle
- BFLH fascicle length
- BFLH volume
- flexibility of hamstrings
- hamstrings muscle stiffness
|
3.4.1 Effect of NHE on the eccentric hamstring strength peak
In the included studies, a total of five experiments from three articles [25,29,33] investigated the effect of NHE on the eccentric hamstring strength peak, with a combined sample size of 76 and 76 for the experimental and control groups, respectively. Heterogeneity tests were performed, and the results are shown in Figure 5, showing no heterogeneity between the combined studies (I2=1%, P=0.04), so a fixed-effect model was used to calculate the combined the results of the effect size showed that NHE significantly enhanced the eccentric hamstring strength peak compared with the control group (MD=46.16, 95% CI: 58.70-63.63, z=5.18, P<0.001).
In the included studies, a total of 11 experiments from 5 articles[23,26,28,30, 34] investigated the effect of NHE on the eccentric peak torque of the hamstrings with a combined sample size of 110 and 95 for the experimental and control groups, respectively. Heterogeneity test was performed, and the results are shown in Figure 6, showing no heterogeneity between the combined studies (I2=0%, p=0.93), so a fixed-effect model was used to calculate the combined effect size, and the results showed that NHE significantly enhanced the eccentric peak torque of the hamstrings compared with the control group (MD=10.87, 95% CI: 3.86-17.89, z=3.04, P<0.01).
3.4.3 Effect of NHE on the eccentric peak torque of the hamstrings muscles to body weight
In the included studies, a total of three experiments from three articles investigated the effect of NHE on the eccentric peak torque of the hamstrings muscles to body weight, with a combined sample size of 38 and 36 for the experimental and control groups, respectively. Heterogeneity test was performed, and the results are shown in Figure 7, which shows that there was no heterogeneity between the combined studies (I2=0%, P=0.82), so a fixed-effect model was used to calculate the combined effect size, and the results showed that NHE significantly enhanced the eccentric peak torque of the hamstrings muscles to body weight compared with the control group(MD=0.17, 95% CI: 0.01-0.33, z=2.02, P<0.05).
3.4.4 Effect of NHE on the hamstrings eccentric angle of peak torque(APT)
In the included studies, a total of 10 experiments from four articles[22,23,27,28] investigated the effect of NHE on the hamstrings eccentric angle of peak torque(APT), with a combined sample size of 103 and 88 for the experimental and control groups, respectively. Heterogeneity tests were performed and the results, as shown in Figure 8,showed no heterogeneity between the combined studies (I2=0%, P=0.76), so a fixed-effects model was used to calculate the combined effect size, and the results showed that NHE significantly reduced the hamstrings eccentric angle of peak torque(APT) compared with the control group(MD=2.53, 95% CI: 4.63-0.43, z=2.36, P<0.05).
3.4.5 Effect of NHE on BFLH fascicle length
In the included studies, a total of four experiments from three articles[26,27,29] investigated the effect of NHE on BFLH fascicle length, with a combined sample size of 54 and 52 for the experimental and control groups, respectively. Heterogeneity tests were performed, and the results were shown in Figure 9, showing no heterogeneity between the combined studies (I2=0%, P=0.57), so a fixed-effects model was used to calculate the combined effect size. The results showed that NHE significantly increased the BFLH fascicle length compared with the control group (MD=0.62, 95% CI: 0.11-1.12, z=2.40, P<0.05).
3.4.6 Effect of NHE on H:Q functional ratio
In the included studies, a total of 3 experiments from 2 articles[23,26] investigated the effect of NHE on H:Q functional ratio with a combined sample size of 28 and 26 for the experimental and control groups, respectively. Heterogeneity tests were performed and the results were shown in Figure 10, showing no heterogeneity between the combined studies (I2=0%, P=0.43), so a fixed effects model was used to calculate the combined effect size. The results showed that NHE training significantly increased the H:Q functional ratio compared with the control group (MD=0.07, 95% CI: 0.02-0.13, z=2.53, P<0.05).
3.4.7 Sensitivity analysis
Sensitivity analysis was performed on the studies included in each research, and none of the studies caused significant interference with the results of this Meta-analysis, implying that this research has a good stability.
3.4.8 Publication Bias Analysis
The software was used for publication bias analysis, and a funnel plot was drawn to check whether there was publication bias in this study. The symmetry of the funnel plot meant that there was no publication bias. See Figure 11-16 for details.