Effect of Active Recovery Protocols on the Management of Symptoms Related to Exercise-Induced Muscle Damage: A Systematic Review

Abstract

Active recovery offers an efficient method to relieve delayed onset muscle soreness and recover from exercise-induced muscle damage (EIMD). The main aim of this systematic review is to identify and compare different active recovery protocols after EIMD. Six databases were searched, and 17 eligible studies were selected. Results showed alleviation of soreness, prevention of muscle strength loss, improvement in flexibility, and a decrease in inflammation after one or more recovery protocols such as isolated muscle contractions, aqua exercise, yoga, and combined jogging and running. A better strategy should focus on prevention of symptoms after EIMD through a precise training periodization and adjustment of load used in exercises.

INTRODUCTION

Physical exercise offers numerous advantages and benefits for both general health status and performance level. For increased effectiveness of exercise, particular precautions and requirements are recommended to prevent the incidence of injuries and speed up the recovery process if an injury occurs. One of the most common and reported occurrences after unaccustomed exercise, known as exercise-induced muscle damage (EIMD), is a phenomenon called delayed onset muscle soreness (DOMS). Delayed onset muscle soreness is defined by a sensation of discomfort experienced in the skeletal muscle and is usually associated with a decrease in muscle force (⁶). The intensity of this discomfort is characterized by an increase within the first 24 hours after exercise, a peak between 24 and 72 hours, a reduction, and finally disappearance by 5–7 days postexercise (⁴⁷). The mechanism underlying DOMS is still unclear, but many hypotheses and theories have been attributed to these phenomena: muscle spasm (⁵⁰), connective tissue and muscle damage (²²), enzyme efflux (¹⁸), hypoxia and ischemia (¹³), and inflammation (¹⁵).

From a physiological perspective, in the absence of muscle pathology, the injury caused by eccentric exercise leads to intramuscular inflammation (³⁸). This reaction is a coordinated dynamic process that leads to adaptive remodeling and a return to homeostasis (³⁹). Blocking or reducing such a response interferes with muscle regeneration and faster recovery (⁹). Particularly, changes in acute phase reactants such as serum C-reactive protein and creatine kinase (CK) as well as increases in circulating white blood cells are observed as a result of inflammation after most types of tissue damage (¹⁶). As a result, prostaglandin and leukotriene synthesis take place; the former being responsible for sensation of pain, and the latter responsible for increasing vascular permeability and attracting neutrophils to the site of damage, generating free radicals, causing swelling in the muscle, and aggravating damage to the cell membrane (^10,11).

The severity of DOMS is influenced by the duration and intensity of EIMD, with intensity being the primary determinant (⁸), although the type of muscle contraction (isometric, concentric, eccentric, or a combination of both) is another factor affecting soreness (²³). Results have shown that isometric and eccentric contractions have caused a higher perception of muscle soreness when compared to concentric contractions, with eccentric causing more soreness than isometric (⁷). Moreover, eccentric contractions compared with rest or other types of contraction have elicited a 30–36% decrease in force (³⁷), 5–7% decrease in range of motion (ROM) (⁵), and an increase in muscle stiffness persisting for several days without significant differences in values of CK plasma levels (^7,14,33,34). A comparison between the effect of running downhill versus running on a level surface on soreness and plasma CK was also investigated and showed that both parameters increased significantly after running downhill (⁴⁴). Moreover, according to a study examining different protocols of high-intensity interval training, short sprints, compared with longer intervals, showed higher muscle damage and muscle soreness (⁵⁵).

In this review, different active recovery protocols were investigated to address DOMS and other associated symptoms. The focus will exclusively be directed on active methods comprising movements or activities performed by the participants. Several studies have favored active over passive recoveries because of potential physiological benefits: better blood lactate removal and increased muscle performance (²⁴), higher muscle voluntary isometric contraction (³⁰), higher total quality recovery, and a decreased feeling of heavy legs (²⁶). The last 2 findings support the use of active recovery not just for the physiological benefits already mentioned but also for positive perceptual and psychological considerations (²⁶). Nonetheless, several active protocols have shown controversial findings regarding the choice of intensity, duration, and type of exercise (^{18,36,41,46,48,51}). To our present knowledge, there are no other reviews that have provided an exclusive and in-depth analysis of active recovery methods to improve performance after EIMD. This review aims at summarizing benefits, identifying limitations, and providing a practical approach for active individuals, coaches, and therapists regarding what type of active recovery could enhance performance level after strenuous activities. The main question to be asked is whether the investigated active protocols meet the demands of sports practitioners in efficiency and offer athletes optimal improvement of their performance level.

METHODS

The methodology adopted in this literature search is implemented according to the guidelines outlined in the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) statement (³¹). The review was registered with PROSPERO (CRD42018114679) at the University of York, United Kingdom. The following sections detail how the review was conducted to address any bias in selection.

SEARCH STRATEGY

A detailed search of 4 major databases was conducted up to and including January 2, 2020. Articles were collected from the following electronic databases: PubMed/MEDLINE, WEB OF SCIENCE, SPORTDiscus, EMBASE, SCOPUS, and CENTRAL. Details of the search strategy for PubMed/MEDLINE are shown in Appendix 1, Supplemental Digital Content 1, https://links.lww.com/SCJ/A307. We also searched clinical trial registries on March 1, 2020, in ClinicalTrials.gov and WHO International Clinical Trials Registry Platform to identify ongoing trials. Additional studies were identified by supplementing the electronic database search with manually screened references of the included articles and citation tracking of included studies in SCOPUS.

ELIGIBILITY CRITERIA

Studies were included if they met the following criteria: trials including experimental and control groups, cross-over study designs, studies including at least one active recovery method, studies including outcome measures taken at different intervals up to a minimum of 48 hours, studies conducted on humans, and articles in English. Systematic reviews or meta-analyses were excluded. After examining the inclusion and exclusion criteria, a total of 17 references were selected for this systematic review (Figure 1).

Figure 1.: Flowchart of the study selection process.

STUDY SELECTION

Search strategy consisted of gathering all results from all databases and removing duplicates using Mendeley Reference Manager. Articles identified by the search strategy were screened independently by 2 reviewers (R.F. and H.O.) using the title and abstract first and then the full text. Disagreements over article inclusion were settled through discussion with a third reviewer until consensus was reached. Data were extracted in duplicate and independently by 2 reviewers (R.F. and H.O.) using an electronic data extraction form. The data extracted included the following: author and year, characteristics of participants (number, training status, sex, and age), and corresponding exercise-induced muscle damage and active recovery protocols.

QUALITY ASSESSMENT

Selected studies were evaluated by 2 reviewers independently (R.F. and H.O.) using a checklist of 5 subcategories and a total of 27 items for assessment of the methodological quality of randomized and nonrandomized studies of health care intervention. The different subscales address, respectively, (a) reporting, consisting of 10 items assessing whether the information provided by the studies gives the reader an unbiased assessment of the findings, (b) external validity, consisting of 3 items addressing to which extent the findings of the studies can be generalized to the population from which the sample size was taken, (c) internal validity (bias), consisting of 7 items for biases in the measurements of the intervention and the outcomes, (d) internal validity (confounding selection bias), consisting of 6 items for biases in the selection of the subjects, and (e) power, to assess whether the negative results of the study are possibly due to chance. A score of 0 or 1 was given to each answer, except for one item in the reporting subscale, which scored 0 to 2, and the last subcategory with one question on power, which was scored 0 to 5. Therefore, the total maximum score was 31. The average score of our selected studies was 17.7 (¹²) (Table1).

Table 1 - Checklist for the assessment of the methodological quality

References	Reporting										External validity			Internal validity—bias							Internal validity—confounding						Power	Score
	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17	18	19	20	21	22	23	24	25	26	27
Andersson et al. (1)	1	1	1	1	0	1	1	0	1	0	0	0	1	0	0	1	1	1	1	1	1	1	1	0	0	1	0	17
Boyle et al. (3)	1	1	1	1	0	1	1	0	1	1	0	0	1	0	0	1	1	1	1	1	1	1	0	0	0	1	0	17
Buroker et al. (4)	1	1	1	1	0	1	1	0	1	0	1	1	1	0	0	1	1	1	1	1	1	1	1	0	0	1	0	19
Gulick et al. (18)	1	1	1	1	0	1	1	0	1	0	1	1	0	0	0	1	1	1	0	1	1	1	1	0	0	1	0	17
Hasson et al. (19)	1	1	1	1	0	1	1	0	1	0	1	1	0	0	0	1	1	1	1	1	1	1	1	0	0	1	0	18
Kawczynski et al. (25)	1	1	1	1	0	1	1	0	1	1	0	0	1	0	1	1	1	1	0	1	1	1	1	0	0	1	0	18
Law et al. (28)	1	1	1	1	0	1	1	0	1	1	1	1	0	1	1	1	1	1	1	1	1	1	1	1	0	1	1	23
Naugle et al. (32)	1	1	1	1	0	1	1	0	1	1	1	1	0	0	0	1	1	1	1	1	1	1	1	0	0	1	0	19
Olsen et al. (35)	1	1	1	1	0	1	1	0	1	1	0	0	1	0	0	1	1	1	0	1	1	1	1	0	0	1	0	17
Sakamoto et al. (43)	1	1	1	1	0	1	1	0	1	1	0	0	0	0	0	1	1	1	0	1	1	1	1	0	0	1	0	16
Takahashi et al. (46)	1	1	1	1	0	1	1	0	1	0	0	0	1	0	0	1	1	1	0	1	1	1	1	0	0	1	0	16
Tufano et al. (48)	1	1	1	1	0	1	1	0	1	0	0	0	0	0	0	1	1	1	0	1	1	1	1	0	0	1	0	15
Wahl et al. (51)	1	1	1	1	0	1	1	0	1	0	0	0	1	0	0	1	1	1	1	1	1	1	1	0	0	1	0	17
Weber et al. (52)	1	1	1	1	0	1	1	0	1	0	0	0	0	0	0	1	1	1	0	1	1	1	1	0	0	1	0	15
Wessel et al. (54)	1	1	1	1	0	1	1	0	1	0	1	1	1	0	0	1	1	1	0	1	1	1	1	0	0	1	0	18
Xie et al. (58)	1	1	1	1	0	1	1	0	1	0	1	1	1	1	0	1	1	1	1	1	1	1	1	1	0	1	1	22
Zainuddin et al. (57)	1	1	1	1	0	1	1	0	1	0	1	1	0	0	0	1	1	1	0	1	1	1	1	0	0	1	0	17

RESULTS

STUDIES INCLUDED

The search strategy and filters were adapted and used for each database to maximize sensitivity and specificity. A total of 2,588 references were researched, of which 1,105 duplicated articles were removed. Titles and abstracts of the remaining articles were analyzed, with irrelevant topics or studies, conference papers, and reviews removed, and the 43 remaining articles were screened for further investigation. After the reading of full-texts copies, 26 articles were excluded and 17 studies meeting the eligibility criteria were included for qualitative analysis (^{1,3,6,20,21,26,31,35,38,43,46,48,51,52,54,56,57}) (Figure 1).

PARTICIPANT CHARACTERISTICS

The qualitative analysis included 481 participants, comprising 190 males and 291 females. Three studies assessed only males (^25,46,51) and 4 assessed only females (^1,3,48,52), and the remaining 10 studies used mixed-sex samples. The number of participants in each study ranged from a minimum of 10 (^19,46) to a maximum of 70 (¹⁸). The average age of participants was 24.4 years, the minimum was 18.7 years (²⁵), and the maximum was 38.0 years (³).

Regarding the fitness level of participants, studies included participants from different types of sports: soccer players (^1,25), long-distance runners (⁴⁶), students practicing different types of sports (⁵¹), yoga practitioners (³), regular recreational physical exercise (³⁵), and also untrained healthy individuals (^{4,18,19,28,32,45,49,52,54,56,57}).

INTERVENTION CHARACTERISTICS

All studies investigated the effect of active recovery protocols after EIMD. Studies were categorized according to the type of recovery protocol and summarized in Table 2. The categories are as follows: isolated muscle contractions (^19,43,57), cycling or arm cranking (^{1,18,35,48,52}), stretching (^4,54,56), general physical activity (^25,28,32), exercise in water (^46,51), and yoga (³). All recovery protocols were performed on the same group of muscles that had induced damage. All studies measured soreness whether through scales, algometer, or questionnaire. Muscle performance parameters such as maximal voluntary contraction (MVC), countermovement jump (CMJ), peak torque (PT), total work (TW), angle of PT, muscle power, stretch-shortening cycle (SSC), sprint time, and dynamic fatigue were measured in 12 studies (^{1,4,18,19,35,45,48,49,51,52,56,57}). Seven studies assessed inflammatory markers: CK (^{1,4,25,45,48,51,57}), myoglobin (^25,51), lactate dehydrogenase (⁵¹), urea, and uric acid (¹). Flexibility was evaluated in 3 studies (^3,46,54), rating of physical exertion was assessed in 3 studies (^3,25,51), active and passive range ROM were measured in 6 studies (^{18,32,43,46,56,57}), muscle circumference was evaluated in 4 studies (^4,18,45,56), and other parameters related to body awareness, disabilities, reaction time, and perceived physical state were assessed in 4 studies (^3,25,33,46).

Table 2 - Different recovery protocols in the management of symptoms after exercise-induced muscle damage (EIMD)

Reference	Subjects (sex, age, groups)	Training status	DOMS-inducing protocol	Active protocol post-EIMD	Outcome measures	Sampling points	Results
Isolated muscle contractions
Hasson et al. (19)	6♂ 4♀ (28.7 ± 8.0) MCE (n = 5) CG (n = 5)	Not specified	10′ bench stepping—15 steps/min	6 sets/20 reps knee F-E	MVC/PT/TW/SPI	PRE, POST, P24H, P48H	Decrease of soreness and SPI, less decrease in muscle performance
Sakamoto et al. (43)	7♂ 5♀ (25.9 ± 3.7) MCE arm (n = 12) Control arm (n = 12)	Untrained active	5 sets/6 reps ECEF	5 sets arm curls until failure (70% MVC), repeated for 4 d	VAS/MC/SJA/CK/MVC/SSC	PRE, POST, P1D, P2D, P3D, P5D, P7D	Increase in static relaxed angle for MC. Increase of circumference, MVC, DEA for control and MC
Zainuddin et al. (57)	10♂ 4♀ (24.4 ± 2.4) LCE arm (n = 14) Control arm (n = 14)	Not specified	10 sets/6 reps ECEF	25′ LCE: 10 sets/60 reps elbow F-E, repeated for 4 d	VAS/tenderness/MVC/CON-PT/ROM (difference between FANG and SANG)/RANG/MC/CK	PRE, POST, P1D, P2D, P3D, P4D, P7D	Decrease of soreness and tenderness immediately after LCE
Cycling and arm cranking
Andersson et al. (1)	17♀ (22 ± 3.4) AR (n = 8) CG (n = 9)	Elite soccer players	One soccer game	20′ cycling (60% HR max), 30′ RT (<50% 1RM) (UE + LE)—10′ cycling (60% HR max)	LIKERT scale/CMJ/ST/PT/CK/UA/U	PRE, POST, P5H, P21H, P27H, P45H, PH51, P69H	No effect
Olsen et al. (35)	15♂ 21♀ (20–30) warm-up (n = 12) cool down (n = 12) CG (n = 12)	Recreationally active	5 sets/10 reps front lunges	Cool down: 20′ MIC, 65–75 rpm (60–70% HR max)	VAS/PPT/MVC	PRE, P24H, P48H	Cool-down group: Decrease of PPT from PRE to P24H
Tufano et al. (48)	26♀ (22.1 ± 2.49) MIC (n = 10) LIC (n = 10) CG (n = 6)	Familiar with DOMS	6 sets/10 reps ECKE	MIC: 20′ cycling, 80 rpm (70% HR max) LIC: 20′ cycling, 80 rpm (30% HR max)	PS/MVC/PT	PRE, POST, P24H, P48H, P72H, P96H	MIC group: Increase in MVC at P72H and P96H from P24H and P72H from IP
Gulick et al. (18)	35♂ 35♀ (21–40) NSAID UEE; IM; SS; OSP; CG	Untrained	15 sets/15 reps ECWE	10′ UEE high velocity (360ᴼ/s)	A + P ROM/MC/volume/VAS/PPT/MVC/TW-PT-angle of PT CON/ECC	PRE, POST, P20’, P24H, P48H, P72H	No effect
Weber et al. (52)	40♀ (18–35) massage (n = 10) UEE (n = 10) NMES (n = 10) CG (n = 10)	Untrained	10 reps or repeated until failure, ECEF, 5''/rep	8′ UEE, 60 rpm, workload 400 Kg-m/min	VAS/MVC/PT	PRE, P24H, P48H	No effect
Stretching
Buroker et al. (4)	16♂ 7♀ (18–33) stretch LKE (n = 7) stretch LKE + RPF (n = 8) CG (n = 8)	Moderate	20′ bench stepping—15 steps/min	10 reps/30 s SS, 10″ rest between reps, left KE and right PF	VAS/PPT/MC/MVC/CK	PRE, P24H, P48H, P72H	No effect
Wessel et al. (54)	13♂ 7♀ (25.2 ± 3.36) stretch pre-EIMD/control leg (n = 10) stretch post-EIMD/control leg (n = 10)	Sedentary	3 sets/20 reps CECKF	10x hamstrings static stretches, 60″ holding,	VAS/PPT/SLR	VAS*: P12H, P24H, P36H, P48H, P60H, P72H/PPT*+SLR*: PRE, P48H	No effect
Xie et al. (58)	20♂ 28♀ (21.7 ± 1.4) DS (n = 16) SS (n = 16) CG (n = 16)	Healthy individuals	3 sets heel raising against elastic band- 120x/min	DS: 10x C-R S + G/5″ IC-30″ holding stretching. SS: 10x SS S + G/30″ holding, 2x/day-5 d	VAS/PPT/ROM/MC/MIC	PRE, POST, P24H, P48H, P72H, P96H, P120H	No effect
General physical activity
Kawczynski et al. (25)	11♂ (18.7 ± 1.2) S1; S2; S3	Professional football field players	3 soccer games	Standard recovery (no activity after the game, 30′jogging at 24h and ball control skills at 48h)/control: No activity recovery/DOMS reduction training (20′ jogging after game, 20′ jogging,, 20′ running-eccentric muscle action related with football and 20′ ball control skills at 24 and 48h)	RPE/PPT/CK/Mb	PRE, P24H, P48H	Increase in PPT for the “DOMS reduction” protocol and increase in CK* for “no activity” and “DOMS reduction session”
Law et al. (28)	23♂ 29♀ (17–40) warm-up + cool down (n = 13) Warm-up (n = 13) Cool-down (n = 13) CG (n = 13)	Not specified	30′ walking backwards downhill, 35 steps/min	Cool down: 10′ walking on inclined treadmill, 4,5-5kph	VAS/NRS/PPT	P10’, P24H, P48H, P72H	Decrease of pain and tenderness in warm-up grp. No significant effect for cool down on soreness or tenderness
Naugle et al. (32)	4♂ 44♀ (20.0 ± 1.9) WB (n = 12) IT (n = 12) LCE (n = 12) CG (n = 12)	Healthy adults	3 sets/10 reps ECEF	20′ WB/8 sets/60 reps EF	Active total ROM/Pain free ROM/VAS/PPT/QuickDASH	PRE, P24, post AR1, P48, post AR2	Increase in EF and pain-free ROM, decrease in VAS and increase in PPT in WB from pretest to posttest. Higher QuickDASH scores for WB at d 2
Exercise in water
Takahashi et al. (46)	10♂ (20 ± 1) AE (n = 5) CG (n = 5)	Long-distance runners	3 sets/5′ downhill running	30′ walking, jogging and jumping in the pool	Stiffness/MP/SR/SL/ROM/CK/WBRT	PRE, P24H, P48H, P72H	Better recovery of soreness in AC. No decline of whole-body reaction time in AC
Wahl et al. (51)	20♂ (24.4 ± 2.2) AC (n = 10) CG (n = 10)	Sports students	300 CMJ, one jump every 8″	30′ LIC in a pool, cadence: 65–75 rpm	VAS/RPE/PEPS/MVC/dynamic fatigue test/Mb/CK/LDH	PRE-EIMD, POST-EIMD, PRE-recovery, POST-recovery, P1H, P2H, P4H, P24H, P48H, P72H	No effect
Yoga
Boyle et al. (3)	24 ♀ (37.8–38.3) YG (n = 12) CG (n = 12)	Yoga trained and no yoga experience	20′ bench stepping (15.5 steps/min)	90′ gentle or moderate yoga class, 24 and 72h after initial testing	BA/SR/VAS/RPE	PRE, P24H, P48H, P72H, P120H	Decrease in pain and increase in flexibility in YG

♂ = male; ♀ = female; MCE = muscle contraction exercise; CG = control group; WB = Wii boxing; IT = ice therapy; AR = active recovery; NMES = neuromuscular electrical stimulation; UEE = upper-extremity ergometry; LKE = left knee extensor; RPF = right plantar flexor; EA = endurance activity; AE = aqua exercise; AC = aqua cycling; S = session; IM = ice massage; SS = static stretching; OSP = ointment sublingual pellets; ECEF = eccentric contractions for elbow flexors; LE = lower extremities; ECKE = eccentric contractions of knee extensors; ECWE = eccentric contractions of wrist extensors, CECKF = concentric-eccentric contractions of knee flexors; F = flexion; E = extension; EF = elbow flexion; LCIE = light concentric exercise; MIAE = moderate-intensity aerobic exercise; RT = resistance training; MIC = moderate-intensity cycling; LIC = light-intensity cycling; HR = heart rate; KE = knee extensors; PF = plantar flexors; C-R = contract-relax; S+G = soleus and gastrocnemius; IC = isometric contraction; DEA = dynamic extension angle; NRS = numerical rating scale; VAS = visual analog scale; A = active; P = passive; MC = muscle circumference; QuickDASH = quick disabilities of the arm, shoulder and hand; SJA = static joint angles; PPT = pressure point threshold; MVC = maximal voluntary contraction; PT = peak torque; TW = total work; CON = concentric; ECC = eccentric; WBRT = whole-body reaction time; SPI = soreness perception index; SSC = stretch-shortening cycle; PS = pain scale; ROM = range of motion; FANG = flexed joint angle; SANG = stretched joint angle; RANG = relaxed joint angle; ST = sprint time; CK = creatine kinase, Mb = myoglobin, LDH = lactate dehydrogenase; UA = uric acid; U = urea; SLR = straight leg raise; STAI = state-trait anxiety inventory; RPE = rate of perceived; CMJ = countermovement jump; PEPS = persons perceived physical state; MP = muscle power; SR = sit-and-reach; SL = stride length; BA = body awareness; MM = magnitude-matching procedure; YG = yoga group.

RESULTS OF RECOVERY PROTOCOLS

After a recovery protocol of knee flexion-extension exercise, a significant decrease of pain and a more modest decrease of MVC, TW, and PT of knee extensors compared with a control group (8.3 vs. 33.4%, 2.3 vs. 13.8%, and 3.8 vs. 12.1%, respectively) were both reported at 48 h (¹⁹). Light concentric flexion-extension exercise of the elbow flexors resulted in a significant immediate decrease in soreness perception (43% on average); however, this effect was short-termed and not sustained (⁵⁷).

Two types of dynamic recovery protocols were performed immediately, 24 h and 48 h after a soccer game: “DOMS reduction training” session (jogging, running, low-intensity eccentric exercises, and ball control skills) and “standard recovery training” session (jogging and ball control skills) (²⁵). Differences in results between before and 48 h after the game have shown a significant increase in pain threshold by 29% after “DOMS reduction training,” whereas a decrease of 19% was noticed after a “standard recovery training” session (²⁵). Plasma CK decreased significantly from 24 h to 48 h after the game by 36.6% when no recovery protocol was performed and by 22.3% after a DOMS reduction training session (²⁵). Naugle et al. have investigated Wii sports boxing active game and a standardized elbow flexion-extension exercise (1lb, 8 sets × 60 reps), both compared with each other and to rest. A greater pain threshold was noticed after Wii boxing along with an increase in active total ROM on day 1 compared with light concentric exercise (150.67° ± 1.16° vs. 148.33° ± 1.16°) (³²).

Exercise in water involving general dynamic exercise for 3 consecutive days (⁴⁶) and cycling (⁵¹) was also investigated. The aqua exercise group showed a significant decrease in calf muscle soreness and stiffness and a significant result on muscle power, where a more pronounced decline was noticed in the control group at day 2 after EIMD (⁴⁶). All blood markers showed a greater increase in the aqua cycling group compared with the control group (⁵¹).

Yoga exercise results showed a lower peak muscle soreness and improved flexibility at 24 h and 48 h after EIMD when compared to the control group (³).

Finally, there were no significant results reported in any of the outcome measures assessed after cycling and arm cranking (^18,35,48,52), combined cycling and low-intensity resistance training (¹), and stretching in both forms, static and dynamic (^4,54,56).

All the outcome measures assessed in the articles are summarized in Table 3.

Table 3 - Different outcome measures used to assess symptoms after exercise-induced muscle damage (EIMD)

References	Outcome measures
	Soreness	Muscle performance	Inflammatory markers	Flexibility	Fatigue	ROM	Other
Andersson et al. (1)	LIKERT scale	PT, CMJ, ST	CK, UA, U
Boyle et al. (3)	VAS			SR	RPE		BA
Buroker et al. (4)	VAS, PPT	MVC	CK				MC
Gulick et al. (18)	VAS, PPT	MVC, TW, PT, angle PT				Active and passive ROM	MC/volume
Hasson et al. (19)	SPI	MVC, TW, PT
Kawczynski et al. (25)	PPT		CK, Mb		RPE
Law et al. (28)	VAS, NRS, PPT
Naugle et al. (32)	VAS*, PPT*					Active total ROM, pain-free ROM	QuickDASH
Olsen et al. (35)	VAS*, PPT*	MVC
Sakamoto et al. (43)	VAS*	MVC, SSC	CK			RANG, FANG	MC
Takahashi et al. (46)	Stiffness by a questionnaire	MP	CK	SR, SL		Total active ROM	WBRT
Tufano et al. (48)	PS	MVC, PT
Wahl et al. (51)	VAS	MVC, dynamic fatigue	CK, Mb, LDH		RPE		PEPS
Weber et al. (52)	VAS	MVC/PT
Wessel et al. (54)	VAS, PPT			SLR
Xie et al. (58)	VAS, PPT	MVC				Active total ROM	MC
Zainuddin et al. (57)	VAS, tenderness by pressure	MVC, CON-PT	CK			ROM (difference between FANG and SANG), RANG

♂ = male; ♀ = female; MCE = muscle contraction exercise; CG = control group; WB = Wii boxing; IT = ice therapy; AR = active recovery; NMES = neuromuscular electrical stimulation; UEE = upper-extremity ergometry; LKE = left knee extensor; RPF = right plantar flexor; EA = endurance activity; AE = aqua exercise; AC = aqua cycling; S = session; IM = ice massage; SS = static stretching; OSP = ointment sublingual pellets; ECEF = eccentric contractions for elbow flexors; LE = lower extremities; ECKE = eccentric contractions of knee extensors; ECWE = eccentric contractions of wrist extensors, CECKF = concentric-eccentric contractions of knee flexors; F = flexion; E = extension; EF = elbow flexion; LCI = light concentric exercise; MIAE = moderate-intensity aerobic exercise; RT = resistance training; MIC = moderate-intensity cycling; LIC = light-intensity cycling; HR = heart rate; KE = knee extensors; PF = plantar flexors; C-R = contract-relax; S+G = soleus and gastrocnemius; IC = isometric contraction; DEA = dynamic extension angle; NRS = numerical rating scale; VAS = visual analog scale; A = active; P = passive; MC = muscle circumference; QuickDASH = quick disabilities of the arm, shoulder, and hand; SJA = static joint angles; PPT = pressure point threshold; MVC = maximal voluntary contraction; PT = peak torque; TW = total work; CON = concentric; ECC = eccentric; WBRT = whole-body reaction time; SPI = soreness perception index; SSC = stretch-shortening cycle; PS = pain scale; ROM = range of motion; FANG = flexed joint angle; SANG = stretched joint angle; RANG = relaxed joint angle; ST = sprint time; CK = creatine kinase, Mb = myoglobin, LDH = lactate dehydrogenase; UA = uric acid; U = urea; SLR = straight leg raise; STAI = state-trait anxiety inventory; RPE = rate of perceived; CMJ = countermovement jump; PEPS = persons perceived physical state; MP = muscle power; SR = sit-and-reach; SL = stride length; BA = body awareness; MM = magnitude-matching procedure; YG = yoga group.

DISCUSSION

The main objective of this systematic review was to examine and assess different active recovery protocols and show potential benefits on the recovery of DOMS and other related symptoms after EIMD. Positive effects on muscle soreness have been shown with isolated muscles contractions but less on muscle performance and ROM. Studies examining cycling for lower and upper extremities as well as static and dynamic stretching did not show significant results on recovery parameters after EIMD. Jogging/running activities showed satisfying results concerning soreness perception and a decrease in inflammatory markers in the blood starting immediately after an intensive activity. Walking uphill seems more efficient when used as a warm-up rather than a recovery protocol with positive effects on the reduction of DOMS. In the same direction, aqua exercise facilitated the recovery of leg muscles and showed faster improvements in muscle soreness, stiffness, and power. Finally, the practice of yoga seems to relieve DOMS and improve flexibility.

Accordingly, strong evidence supports the effect of muscle contractions on soreness. Adding motivation and fun to exercise to maximize its effects compared with other active methods has also been revealed to be very important (³²). Similarly, the choice of the intensity of exercise might influence the present results; hence, Wii boxing involving light-to-moderate intensity might lead to higher muscle activation and can potentially be required to optimize the effectiveness of exercise in the days after the intervention protocol (³²). Conversely, the use of light concentric exercise (LCE) involving repetitions of flexion-extension without weight, performed in the days after an intensive training session has shown a positive result on soreness only immediately after the recovery protocol (⁵⁷). This finding correlates with the “exercise-induced analgesia” hypothesis which showed that light-to-moderate intensity seems to contribute in alleviating soreness, although the effect is temporary and not sustainable (²⁷). Therefore, from a practical perspective, LCE seems to be recommended to help to relieve soreness immediately after an intensive training session.

A possible explanation for the lower decrease in muscle performance after knee flexion-extension is the potential effect that high-speed maximal contractions (300°/sec) or shorter time under muscle tension have on inflammatory markers (¹⁹). During eccentric and high-speed contractions, predominantly fast-twitch units are recruited (²⁰), thus stimulating type II muscle fibers that potentially are more often injured than type I (⁴⁰). The selective recruitment of these affected type II fibers reduces the fluid accumulation and swelling and delays the upregulation of proinflammatory cytokines such as nerve growth factor, histamine, and prostaglandins that are responsible for nociceptor activation (³⁴). Therefore, participants showed a better muscle performance in the exercised leg because of the reported decrease in perceived pain compared with the control leg. However, upper extremities were studied (^32,43,57), and no significant difference was found in all muscle performance-related parameters between the exercising and control groups. The studies using upper extremities stated no significant positive result on muscle recovery after a damaging exercise and fatigue. The reason behind that might be related to the muscle mass involved in the recovery as the lower extremity has a greater muscle mass than the upper extremity; as a result, the exerted muscle contraction is more accentuated and might lead to significant positive findings.

The contradictory results between the 2 recoveries on ROM could be attributed to the difference in LCE protocols performed (^32,57). For example, 10 sets of 60 reps (2 sec/flexion-extension), using an isokinetic dynamometer at 240°/s, were performed for 20 minutes in one study (⁵⁷), whereas 8 sets of 60 repetitions of flexion-extension with a free weight of 1 lb. were included in another study (³²). This difference in the mode of exercise could be translated into a modification in muscle tension occurring between isotonic and isokinetic exercise for speed velocities and load intensities. Recently, some authors have highlighted the importance of speed to quantify the intensity of an exercise, leading to differences in performance (^37,41).

Neither the upper or lower extremity showed a significant change in pain or on isometric or dynamic strength after the application of a low or moderate intensity protocol (^1,18,25,35); however, an increase in isometric strength at 72 and 92 hours compared with 24 hours post-EIMD was found in one study after moderate cycling with no effect on dynamic strength (⁴⁸). The result is in contrast to previous findings of previously mentioned studies stating the positive effect of isolated muscle contractions on soreness perception (^19,43,57). The reason behind this dissimilarity might be attributed to the choice of intensity in the examined recovery protocols. Acute endurance exercise is related to higher blood flow and an increase in muscle perfusion helping in removing noxious waste products after a high-intensity exercise causing contractile elements disruption (⁴⁹). It was suggested that to increase this muscle perfusion and thus decrease soreness, heart rate should be elevated before each exercise to enhance tissue repair (¹¹). Consequently, the light endurance intensity (between 30 and 60%HRmax) used in cycling and arm cranking (^1,48) and the submaximal contractions in upper ergometry (360/s) (¹⁸) might not have been effective in obtaining a considerable decrease in muscle pain, whereas higher intensities (between 60 and 70% HRmax) have shown better results on muscle performance (⁴⁸).

In addition, no significant difference was detected after a stretching protocol on pain and soreness perception. This is consistent with all the studies involved in this review (^4,54,56). The soreness outcomes are reported to reflect the process of muscle nociception in response to the accumulation of endogenous analgesic agents (¹⁷). The fact that participants cannot be blinded during their recovery protocol is an important factor to be mentioned because subjective positive perception was expected in the groups that performed stretching, which leads to a greater expectation of relief after their session (⁵⁴). Although dynamic stretching was suggested for its reported analgesic effect caused by isometric contraction and thus interrupting the transmission of pain (^21,45), the results did not support this idea. In addition, there was no positive effect observed on muscle performance and inflammatory markers, which might be due to the low level of muscle damage reported after a combination of concentric and eccentric contractions instead of pure eccentric bouts to induce muscle deficits soreness (⁵⁶).

General physical activity involving warm-ups and cool downs showed potential benefits including increased temperature and muscle compliance, reduced muscle strain injuries (⁴²), and decreased DOMS (²). In the reviewed study, only warm-up showed a significant effect at 48 hours after eccentric exercise on reducing soreness and muscle tenderness (²⁸). This finding justifies the use of dynamic exercise before the main activity to alleviate soreness through the removal of waste products by stimulating blood flow and increasing endorphins (⁸).

Consequently, the main difference with all of the previously covered studies is the inclusion of a high-volume exercise protocol performed 24 and 48 hours after an intense activity, such as a soccer game. This extra stimulation to the main muscle affected by DOMS may lead to a higher level of interference in the pain sensation mediated by mechanoreceptor fibers (⁵³).

Regarding the positive effect on plasma CK activity, an immediate active recovery protocol seems to help decrease blood markers after activity while also increasing blood circulation and removing metabolic waste products, such as carbon dioxide, urea, and uric acid. This may explain the decrease in CK seen after implementing an immediate active protocol with a bigger volume (²⁵).

Exercising in water has also been found as an effective method for dealing with soreness (⁴⁶). A possible explanation for this could be the decrease of load over the legs when immersed into water, causing a massage effect which, therefore, could increase peripheral blood flow, facilitate the elimination of edema, and reduce local muscle stiffness. Consequently, muscles seem to be less swollen and less painful. This massage effect could be lower when performing cycling in water (⁵¹), which may explain the similarity of results on pain intensity following that modality and in the control group.

Nevertheless, the increase in blood markers after aqua cycling might be due to the effect of a localized exercise (cycling) performed on the anterior thigh muscle in water, after a highly fatiguing activity (300 jumps), which might have potentially accentuated the muscle damage (⁵¹). The positive effect of aqua exercise on muscle performance was also supported in previous studies showing the benefit of water jets in preventing any potential loss of muscle power (⁴⁹).

Finally, results on yoga practice showed positive results on muscle soreness. This could be related to originally existing differences in muscle condition for yoga practitioners in comparison with untrained participants (³). Flexibility was also improved and could explain the improvement in pain intensity after the bouts of yoga because flexibility has always been linked with a decreased perception of muscle soreness in yoga practice (²⁹).

CONCLUSION

Active recovery protocols showed several advantages and benefits over rest with the main improvements observed on the magnitude of soreness. Muscle isolated contractions, jogging, running, aqua exercise, and yoga all seem to offer limited management of DOMS after EIMD. Aqua exercise prevented to some extent a loss in muscle power, jogging and running decreased inflammatory markers, and the practice of yoga improved flexibility. On the other hand, arm cranking, cycling, and stretching did not seem to help soreness or any other symptom. Furthermore, the impact of active recovery was less obvious on outcomes related to functional capacity and performance. It is also important to note that none of the tested methods have shown an adverse effect on soreness. Consequently, the attempt of decreasing the recovery period without affecting an athlete's overall performance is still a critical challenge and requires further investigations to optimize training benefits and outcomes.

PRACTICAL IMPLICATIONS

This review offers a few practical implications for sports practitioners. Although no particular active recovery was found to be efficient in helping the improvement of performance level after an intensive training session, it is essential to highlight the benefits on different aspects related to performance. Exercise under water might be a good option to limit the decrease in muscular strength after an intensive training session. In addition, pain due to soreness can be relieved by several recovery protocols ranging from isolated muscle contractions to general physical activity.