When to Start Plyometric Training After Anterior Cruciate Ligament Reconstruction? A Systematic Review of Randomized Clinical Trials

Eligibility Criteria

This systematic review includes studies on rehabilitation after ACLR in an adult population. Only studies published in the English language in peer-reviewed journals were considered. The language restriction was based on previous studies showing that when non-English publications were excluded from evidence syntheses, similar conclusions were reached.^34,36 Only randomized controlled trials were included (highest level of evidence) in the study, and the outcome measures contained strength, pain, range of motion, patient-reported outcome measures, swelling, laxity, functional activities, adverse events, and return to activity.

Participants with musculoskeletal injuries (such as ankle or hip injuries) were excluded from the study. In addition, studies involving pediatric interventions after ACLR, as well as those using animals, cadavers, or laboratory-based experimental models, were excluded. A variety of studies were also excluded, including systematic reviews, meta-analyses, nonrandomized trials, observational studies, case reports, medical record analyses, narrative reviews, editorials, and letters. Studies that did not report relevant outcomes or that focused exclusively on concomitant injuries (e.g., meniscal rehabilitation) were also excluded.

Search Strategy

PubMed, Cochrane Library, and PEDro databases were searched to find pertinent literature relevant to the study from inception to March 13, 2026. The search was performed in the database with keywords (search criteria used in each database is available as Supplemental Material in the online version of this article) and Boolean operators were used to identify relevant literature. Medical terms (MeSH) and search manager were used to find studies in a more comprehensive and precise manner.

Study Selection

Potentially eligible articles were selected independently by 2 reviewers. Each reviewer independently screened for eligibility all titles and abstracts (using EndNote X9) of all randomized controlled trials that were identified. For the final selection of studies, 2 reviewers independently reviewed the full texts of all potentially relevant studies. Whenever there was a disagreement between 2 reviewers, a third reviewer was consulted to reach an agreement. In addition, the references from selected studies were read and explored to find more related studies.

Quality Assessment of Studies

An updated version of the Cochrane tool for the assessment of the risk of bias in randomized trials (RoB Version 2.0 [ROB2]) was used for risk of bias assessment. ⁵² The ROB2 Excel sheet was independently used by 2 reviewers. A signaling question was used to identify 5 domains of bias concerns. The overall signaling questions were then used to determine the overall bias risk (high, low, some concern). A third reviewer was consulted about debates in the initial risk of bias ratings until an agreement was reached.

To assess the quality of the body of evidence, the Grading of Recommendations, Assessment, Development and Evaluation (GRADE) approach was used. The first step was to determine and classify all potentially important patient outcomes in accordance with the GRADE guidelines (details in Supplemental Material, available online). A set of criteria for assessing the quality of evidence was also used. The certainty of the evidence for each comparison and outcome was rated as high, moderate, low, or very low based on the risk of bias, inconsistency, indirectness, and imprecision. ⁴⁷ As part of the publication bias assessment, a funnel plot was to be used for meta-analyses that included ≥10 trials. ²⁴ An online GRADEpro Guideline Development Tool (https://www.gradepro.org/) was used to assess the body of evidence and create summary tables.

Data Extraction and Collection

A specific form, which includes author, study design, population (sample size, sex, and age), groups, type of graft, period of intervention, time for starting plyometric, criteria for return to plyometric training, outcome measure of study and adverse event, was designed to collect data from the literature for answering the current question. During the data extraction and collection phase, 2 reviewers worked independently from each other and consulted a third reviewer when necessary to decide each study’s quality and eligibility. The extraction form was pilot-tested before extracting information from the articles. The information obtained from data extraction was summarized primarily in a table as a qualitative synthesis for this systematic review. In addition, a meta-analysis (quantitative synthesis) was to be conducted when ≥2 studies met the criteria for clinical homogeneity. ¹³

Data Items

Data were extracted for all predefined outcome measures of interest. No restrictions were imposed on the measurement instrument or scale used in the eligible studies, or on the duration of follow-up, and outcome data were collected at all reported timepoints.

Statistical Analysis

To conduct the meta-analysis, Review Manager (RevMan), developed by the Cochrane Collaboration, was used. In healthcare research, RevMan supports rigorous systematic review preparation and statistical analysis. In addition, it allows structured data input, analysis of findings using fixed-effects and random-effects models, and generation of outputs such as forest plots. Furthermore, the software facilitates the assessment of between-study heterogeneity (e.g., using the I² statistic) and the evaluation of risk of bias in compliance with established methodologies. The latest version of RevMan Web can be accessed via the official Cochrane website (www.cochrane.org). RevMan Web is a web-based platform that facilitates collaborative workflow, and it provides advantages such as automatic updates, secure data management, and seamless integration with other Cochrane resources. This enhances efficiency, transparency, and reproducibility of the review process.

Study Selection

Based on a database search of relevant literature reviews, 17,860 reports were identified as potentially useful for the study. To remove duplicate reports, EndNote X9 software (n = 1633) was used; thereafter, titles and abstracts of the remaining unique reports (n = 16,227) were screened. A total of 93 full reports were retrieved and assessed for eligibility. Of these, 16 studies were included and reported in the systematic review (Figure 1).

OPEN IN VIEWER

Figure 1.

Flowchart for the systematic review.

Study Characteristics

Detailed descriptions of the 16 included studies are provided in the Supplemental Material, available online.^{3,5,9,10,16 -18,25 -29,44,45,48,51} The total sample size was 685 participants, of which 70% were men and 30% were women. Participant mean age was 25.23 (±4.34) years. The grafts used for the majority of the operative interventions (80%) for the participants were hamstring tendon autograft (HS; 45.40%) and bone-patellar tendon-bone (BPTB; 34.60%). In the remaining participants, different types of grafts were used, including allografts and autografts (15.04%), tibialis anterior tendon allografts (4.67%), and cadaveric grafts (0.29%). However, in 13 studies, functional training exercises incorporating plyometrics were compared with conventional rehabilitation programs.^{5,10,16,18,25 -29,44,45,48,51} One study compared the plyometric training exercises (PTE) with a combination of plyometric and a perturbation program. ³ The effect of low-intensity versus high-intensity PTE was compared in 1 study. ⁹ Finally, 1 study compared the effect of PTE under normal bodyweight with PTE augmented with bodyweight support. ¹⁷

Risk of Bias and Certainty in Studies

The majority of study outcomes (63.5%) were evaluated with some concerns, while the remaining outcomes had either a low (17.3%) or high (19.2%) risk of bias for overall assessment (Figure 2). Detailed risk of bias assessments is provided for all outcome measures (in Supplemental Material, available online). In total, 10 outcome measures from 5 studies ^{5,25,29,45,48} were judged to have high risks associated with deviations from the intended intervention domain. This was related to 5 different measurement domains as follows: patient-reported outcome measures in 1 study assessed high risk, ⁵ functional performance measures in 3 studies,^5,25,29 strength outcome measures in 2 studies,^29,48 knee laxity measurement in 2 studies,^5,48 pain outcome measures in 1 study, ⁴⁵ and proprioception outcome measures in 1 study. ²⁹ The main reasons for judging “high risk” were that appropriate analysis was not performed (intention-to-treat),^{5,25,29,45,48} no detailed information was provided regarding blinding or randomization process,^5,29,45,48 and pain measurements were different from those intended. ⁴⁵

OPEN IN VIEWER

Figure 2.

Risk of Bias Assessment Using the Cochrane RoB2 Tool. RoB2, risk of bias in randomized trials Version 2.0.

In addition, 3 outcome measures were deemed to have high risks associated with missing outcome data. There was 1 study with high risk that examined functional performance measures, strength outcomes, and proprioception outcomes. ²⁹ This judgment was made due to missing outcome data, which was less than the accepted level of 95%, which could be the result of participants’ health conditions.

All certainty assessment results were detailed in the summary of findings tables (details in Supplemental Material, available online). The majority of study outcomes included in this systematic review showed very-low to low certainty results for the critical outcome. In contrast, the important outcome was very-low to moderate.

Specific Criteria to Start PTE

This report included 10 studies (63%) using specific criteria to determine when patients should begin plyometric exercises after an ACLR (Table 1),^{3,9,17,18,25,29,44,45,48,51} whereas in 6 studies (38%),^{5,10,16,26 -28} the plyometric exercise program was carried out based on a timeframe for the healing process. In 6 studies (60%) muscle strength measurements were used as a criterion for starting PTE.^{3,9,18,25,48,51} Table 1 illustrates the different types of specific criteria domains for starting PTE, as well as the number and percentage of studies and outcome measures for each criterion.

OPEN IN VIEWER

Table 1.

Specific criteria to start PTE after ACLR

Type of criterion	Outcome measurement	Number of studies (%)
Muscle strength criteria		6 (60%)
Isokinetic dynamometer
	Isokinetic concentric peak torque difference ≤15% in knee flexor-extension moment at 180 deg/s and 300 deg/s 48
	Isokinetic concentric peak torque difference ≤15% in hip abduction-adduction moment at 60 deg/s and 120 deg/s 48
	Isokinetic deficit of <70% on the contralateral side with symmetry 51
	Up to 70% of isokinetic muscle force 25
Hand-held dynamometer
	80% of quadriceps strength limb symmetry 18
	Not specified tools of outcome measure
	Minimum 80% quadriceps strength limb symmetry 3
	Quadriceps index ≥60% 9
Clinical based criteria
Effusion or swelling		6 (60%)
	Minimal effusion 3
	No effusion or swelling18,25,51
	Swelling amount44,45
Range of motion		5 (50%)
	Full range of motion3,25,51
	Full active knee extension 9
	Active knee flexion within 5 degrees of the contralateral side 9
	Knee range of joint motion determined by electro goniometer 18
Pain		5 (50%)
	No pain or pain-free knee active range of motion3,18
	Pain rating ≤2 out of 10 during activities of daily living 9
	Pain level44,45
Functional performance-based and balance criteria
Functional performance		3 (30%)
	Complete single leg forward hop, triple hop, crossover hop, and 6-meter timed hop without pain and ≥80% of the contralateral limb 48
	1-legged hop test 29
	Single-legged hop for distance limb symmetry index <75% 17
Balance		2 (20%)
	Ability to maintain static and dynamic balance44,45
Biomechanics-based and patient-reported outcome criteria
Biomechanics		2 (20%)
	Double leg squat on a force plate with a difference of 20% of the contralateral limb 48
	Symmetry of 10 consecutive tuck jumps 48
	Vertical ground reaction forces with ≤10% difference during performing single leg drop jump task and ≤3 times body mass load 48
	Peak knee moment during single-legged landing <2.3 bodyweights and <80% of the contralateral limb 17
Patient-reported outcome		1 (10%)
	International Knee Documentation Committee Subjective Knee Form questionnaire with scores of <7517
Other criteria		3 (30%)
	Completing a running progression 3
	Awareness of body position44,45

ACLR, anterior cruciate ligament reconstruction, PTE, plyometric training exercises.

Strength-Based Criteria

Strength was assessed by isokinetic dynamometry in 3 studies,^25,48,51 hand-held dynamometry in 1 study, ¹⁸ and 2 other studies did not report using tools to collect the outcome data of the strength criteria.^3,9 One study reported using isokinetic concentric peak torque difference ≤15% for knee flexor-extension moment at 180 deg/s and 300 deg/s and hip abduction-adduction moment at 60 deg/s and 120 deg/s. ⁴⁸ One study reported using the isokinetic outcome measure with a criterion of a deficit <70% for symmetry with the contralateral side. ⁵¹ One study adopted an isokinetic outcome measure, using a threshold of ≤70% isokinetic muscle force. ²⁵ One study reported using a hand-held dynamometer with a criterion of 80% quadriceps strength limb symmetry to clear patients for starting plyometrics training exercises. ¹⁸ Two studies reported using quadriceps strength limb symmetry index of 80% (determined by dividing the maximum torque of an involved limb by the maximum torque of an uninvolved limb × 100%) and ≥60% (the calculation by utilizing a normalized peak knee extensor torque [peak knee extensor torque/body mass in kg] and the quadriceps index (peak knee extensor torque on the surgical side/peak knee extensor torque on the nonsurgical side × 100)).^3,9

Clinical-Based Criteria

Three studies reported using a criterion of no effusion or swelling to begin PTE.^18,25,51 Arundale et al ³ reported minimal effusion as a criterion, and 2 studies reported using a swelling amount measure for starting PTE.^44,45 A criterion of complete knee range of motion was used to initiate PTE in 3 studies.^3,25,51 Chmielewski et al ⁹ reported full active knee extension and active knee flexion within 5 degrees of the nonsurgical side as a criterion to clear the patients. Ghaderi et al ¹⁸ reported only the knee range of joint motion via an electrogoniometer without a specific range limit. There were 2 studies that used the criterion of no pain or pain-free active range of motion in the knee to start plyometric training.^3,18 In the study by Chmielewski et al, ⁹ a pain rating of ≤2 out of 10 during daily activities was used as a criterion. Two studies used pain level to indicate when to begin plyometric training without mentioning a specific level limit.^44,45

Functional Performance-Based and Balance Criteria

Ghaderi et al ¹⁸ reported using a single-leg forward hop, triple hop, crossover hop, and 6-meter timed hop at an equivalent distance and/or rate of ≥80% of the contralateral side. Elias et al ¹⁷ also reported using a single-legged hop for a distance limb symmetry index <75% to indicate when to begin. Kaya et al ²⁹ reported using a single-legged hop test to determine the functional level of patients without providing a specific capacity limit or cutoff as a criterion. In 2 studies,^44,45 it was reported that a patient’s ability to keep a static and dynamic balance was used as a criterion to determine when it was time to begin plyometric training without providing a specific cutoff point.

Biomechanics-Based and Patient-Reported Outcome Criteria

Setuain et al ⁴⁸ reported using multiple criteria to clear patients for training, including load distribution symmetry ≤20%, qualitative tuck jump symmetry, vertical ground reaction force difference ≤10% during a 50-cm single-leg drop jump while maintaining a load of ≤3 times body mass. Elias et al ¹⁷ reported using a peak knee moment while performing a single-legged landing with <2.3 bodyweights and <80% of the nonsurgical side. However, only 1 study reported using the International Knee Documentation Committee Subjective Knee Form questionnaire with scores of <75 as a criterion for introducing plyometrics training exercises. ¹⁷

Time to Start PTE

A total of 14 studies reported plyometric training starting times with a mean of 13.50 weeks and a standard deviation of 3.03 (mean ± SD = 3.38 ± 0.76 months) (Figure 3).^{3,5,9,10,16 -18,26 -29,44,45,51} In 2 studies,^25,48 no information was reported about when to begin PTE. Table 2 demonstrates the frequencies and categorized distribution of weeks at which plyometric training was initiated after ACLR. However, in 8 studies that used specific criteria to clear patients, PTE were started after a mean of 14.75 weeks and a standard deviation of 3.15 (mean ± SD = 3.69 ± 0.79 months).^{3,9,17,18,29,44,45,51} In 6 studies that used a timeframe,^{5,10,16,26 -28} PTE were started after a mean of 11.83 weeks and a standard deviation of 2.04 (mean ± SD = 2.96 ± 0.51 months).

OPEN IN VIEWER

Figure 3.

Illustration of starting time of PTE after ACLR. ACL, anterior cruciate ligament; ACLR, ACL reconstruction; PTE, plyometric training exercises.

OPEN IN VIEWER

Table 2.

PTE starting time frequencies and categories after ACLR

Start week	Start month (approximately)	Number of studies	Studies
9	2.25	1	Jauregui Bidegain et al 25
10	2.5	1	Jin et al 26
12	3	6	Arundale et al, 3 Beynnon et al, 5 Chmielewski et al, 9 Christensen et al, 10 Elias et al, 17 Ghaderi et al 18
14	3.5	2	Kasmi et al27,28
16	4	1	Souissi et al 51
17	4.25	2	Risberg and Holm, 44 Risberg et al 45
20	5	1	Kaya et al 29
Category	Range of start week	Observation	Studies
Very early start	≤10	Uncommon	Jauregui Bidegain et al, 25 Jin et al 26
Early start	12	Most commonly used	Arundale et al, 3 Beynnon et al, 5 Chmielewski et al, 9 Christensen et al, 10 Elias et al, 17 Ghaderi et al 18
Mid start	14-16	At transitional stage approximately	Kasmi et al,27,28 Souissi et al 51
Late start	≥17	More cautious approach	Kaya et al 29 Risberg and Holm, 44 Risberg et al 45

ACLR, anterior cruciate ligament reconstruction, PTE, plyometric training exercises.

Length of PTE Intervention

Training interventions varied between studies from 4 weeks to 12 weeks in terms of the duration of plyometrics training. One study implemented PTE for 4 weeks, ²⁹ 1 study for 5 weeks, ³ 2 studies for 6 weeks,^27,28 3 studies for 7 weeks,^5,44,45 3 studies for 8 weeks,^17,18,51 2 studies for 12 weeks,^9,10 and 4 studies did not report the length of training.^16,25,26,48 The most common duration range was 7 weeks to 8 weeks for the PTE after ACLR.

Adverse Events Associated With PTE Intervention

Only 2 studies reported that patients were not adversely affected by PTE when performed as part of their training program.^3,28 The other studies did not provide information as to whether or not there were adverse events that occurred during the training period for the patients with ACLR.^{5,9,10,16 -18,25 -27,29,44,45,48,51}

Statistical Syntheses Results

None of the included studies investigated the criteria for starting plyometrics training exercises and none provided data related to the included criteria. Therefore, conducting a meta-analysis to explore the efficacy of the criteria based on returning to plyometrics training exercises after ACLR was impossible. Thus, since there were not 10 trials to include in a funnel plot, publication bias was not assessed and sensitivity analysis was also not possible. The effectiveness of the plyometrics training exercise was not investigated because it was outside the scope of this systematic review.

Clinical Implications

Advanced rehabilitation exercises involving plyometrics exercise improved knee function and impairments as well as boosted psychological wellbeing. The time to start plyometrics exercise has no consensus timeframe. Studies reviewed demonstrate that patients with ACLR have been initiated on a plyometric training program at an average of 13+ weeks. A criteria-based assessment with a concomitant timeframe after ACLR might be a reasonable approach for practitioners. The criteria as a precursor to initiate PTE include no effusion, full active knee extension with active knee flexion within 5 degrees of the nonsurgical side, no pain or ≤2 out of 10 pain rating during daily activities, and the International Knee Documentation Committee Subjective Knee Form questionnaire with scores of 75 can be a prerequisite for introducing plyometric training. Before starting plyometrics training exercises, a battery of tests may also be considered, such as strength-based criteria, performance-based criteria (functional and balance), and biomechanics-based criteria.

Nevertheless, each patient with ACLR should be assessed individually before returning to plyometric exercise for safety purposes. Some measurements may be more pertinent for certain patients with ACLR, depending on their sport performance profile. Researchers and practitioners should consider the patient’s individual sport when designing research and treatment plans.

Study Limitations and Future Research

This systematic review study has limitations that need to be acknowledged. Only randomized control trials were included; other study designs were not included, which may provide different valuable information. The study design restrictions were implemented to minimize bias and promote evidence certainty. Different types of grafts that have been used for the ACLR in the included studies may influence the time to start PTE. The presence of concurrent injuries, such as meniscal injuries with ACL injury, is likely to have additional healing concerns that could prolong the return to PTE. ²⁰ In the case of revision ACLR, more substantial chondral damage can adversely affect functional outcomes and activity levels. ⁵⁵

Consequently, the time of returning to PTE may need to be changed based on the healing process requirement. A patient’s psychological condition can influence their ability to return to sport after ACLR, ³⁷ but no study included here used psychological outcome measures as a criterion to initiate plyometric training. However, the systematic review search for relevant literature was completed on March 13, 2026, which may not reflect newly developed studies thereafter. More research is required to validate the proposed assessment criteria for starting plyometric training safely with an appropriate equilibrium between loading and graft healing after ACLR. It is also important to investigate whether psychological status affects the readiness of patients to handle the load associated with advanced rehabilitation. In addition, future research may consider different study designs to extend insight regarding the time of starting plyometric, parameter (load, specificity, and frequency), and safety to optimize the related load of exercise post-ACLR.