Acute effects of the Prevent injury and Enhance Performance warm-up on potential reinjury risk factors associated with unanticipated jump landings: Study protocol for a crossover randomized controlled trial in individuals cleared for return to sports after anterior cruciate ligament reconstruction.

In team-sports such as football or basketball, athletes need to rapidly adapt their motor plans and actions to unanticipated changes in the environment. Unanticipated jump-landing tasks have been found to elevate the risk of non-contact anterior cruciate ligament (ACL) injuries compared to an anticipated condition. ACL-reconstructed individuals may have greater diculties to maintain neuromuscular control under unanticipated conditions exposing them to a higher reinjury risk during the game. The planned trial aims to investigate the acute effects of a team-sport specic injury prevention programme on potential ACL reinjury risk factors under anticipated and unanticipated jump-landings. and design: movement (equally informed unanticipated participants will the same (consistent with or conditions, and

performance enhancing approaches. [5][6][7] Since primary ACL injury prevention training programmes have been successful at reducing the risk of ACL injuries. [8] The transfer of such primary programmes to secondary injury prevention may be promising.
The use of preventive strategies with a focus on mechanisms of ACL graft rupture mostly include warmup exercises with dynamic single-leg jump landing manoeuvres, plyometrics, agility and neuromuscular strength enhancing activities. [9][10][11] Beyond strength and hop tests having predictive value for reinjury after ACL reconstruction (ACLR), [12] fast motor reaction/decision-making to unpredictable external signals may be another important reinjury risk predictive factor. During sports participation, athletes are required to rapidly process a multitude of visual stimuli (e.g. location and movement of ball, teammates and opponents) while simultaneously monitoring and adjusting their own motor plans and actions to unforeseen demands from the environment. [13,14] Due to time constraints, decision-making within an extremely short time period during athletic movements (e.g. landing from a jump) does not often allow for su cient feed-forward muscle contraction. [4,15,16] Accordingly, Almonroeder et. al. [17] found that these unanticipated affordances result in reduced biomechanical landing safety compared to anticipated tasks, where the stimulus indicating movement direction is shown prior to movement thereby increasing the odds of sustaining a non-contact ACL injury. In addition, unanticipated trials have been found to predispose athletes to wrong and prolonged landing-related decision-making (e.g. landing on the wrong side). [18][19][20] According to a previous study, [20] these decreases of biomechanical landing safety and decision-making quality under the unanticipated relative to anticipated can be de ned as 'unanticipated landing costs'. These unanticipated landing costs may contribute to an increase in the residual risk for reinjury following RTS. Based on the reported high ACL reinjury rates, [2,3] one may speculate that ACLrecostructed individuals have greater di culties in dealing with unanticipated circumstances.
In fact, Clark et. al. [21] found that ACL-reconstructed individuals exhibit altered landing mechanics during drop-jumps with unanticipated side-cuttings compared to an uninjured control group. Niederer et. al. [22], found that ACL-reconstructed patients show impairments in unanticipated jump landing movements for at least 18 to 26 months after surgery. This time span ts the proposed high-risk-for-a-second-ACL-injuryperiod [2,3] after ACLR and RTS. Niemeyer et. al. [23] also reported that performing unanticipated single leg jump landing tasks following RTS clearance after ACLR provides additional information about functional restoration than classic RTS related hop-tests. These de cits and additional information areinvisible during anticipated landings since it is characterised by su cient time for feed-forward control. The unanticipated conditions seem to reveal potentially persistent postoperative neuromuscular de cits leading to an increased reinjury risk on RTS.
Since, most injuries occur during a match or other competitive situation requiring time constrained decision making ACL-reconstructed athletes cleared to RTS may bene t in particular from the acute effects of injury preventive protocols aiming to improve neuromuscular performance. A promising neuromuscular preventive warm-up programme which may be used as a secondary preventive approach for ACL reinjury may be the Prevent Injury and Enhance Performance (PEP) programme. It was designed to target ACL injury risk factors and is able to strongly reduce the incidence risk of primary ACL tears up to 82%. [24]. As most of the preventive programmes, the PEP programme is utilized as a warm-up strategy. Several studies [4,[25][26][27][28] investigated the acute effects of such warm-up programmes (in heathy participants) and a hypothetical positive impact on potential reinjury risk factors under both anticipated and unanticipated conditions in healthy individuals were found.
In ACL-reconstructed individuals, however, the potential acute effects of ACL-speci c neuromuscular warm-up strategies on unanticipated jump landings ability are yet to be determined. Hence, the planned trial aims to determine the acute effects of the PEP warm-up protocol on potential (1) biomechanical reinjury risk factors during both anticipated/pre-planned and unanticipated/un-planned landing conditions as well as (2) unanticipated decision-making quality in ACL-reconstructed participants cleared for RTS. We hypothesise that an injury preventive warm-up programme (PEP) leads to improved jump landing stability and decision making quality in unanticipated relative to anticipated landings resulting in lower unanticipated landing costs.

Study design
This study is a single centre crossover randomized control trial. It follows the Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) [29] and its accompanying explanation and elaboration [30] as depicted in the SPIRIT ow diagram in Fig. 1

Ethics and informed consent
Ethical approval was obtained from an independent Ethics Committee of the Goethe University, Frankfurt (Main), Germany (Faculty of Psychology and Sports Sciences). All prospective participants will be provided with two copies of written informed consent to sign and retain one after familiarising themselves with the relevant study participants' information.

Sample size determination
No study to retrieve the minimal clinical relevant change for the primary outcome during a single leg landing counter movemtn jump (CMJ) following an injury preventive warm-up programme on anticipated and unanticipated landing mechanics in ACL-reconstructed participants has been published yet. Therefore, our a priori sample size calculation (G*Power software version 3.1.9.2; Universität Düsseldorf) was based on standard assumptions. We considered a two-sided alpha-error (α) level of 5%, a target effect size of 0.5, and an 80% power. Accordingly, we will need data from 19 ACL-reconstructed participants (including an anticipated 10% drop out rate).

Recruitment
Participants' recruitment will be done primarily by contacting local surgeons and rehabilitations specialists, sports rehabilitation centres, physiotherapy and medical practices, sports clubs, tness centres and the sports and medical campuses of the Goethe university in Frankfurt (Main), Germany by means of a poster with detachable contact information, yers, telephone calls, e-mails as well as personal addressing. Additional recruitment will consist of advertisements on relevant professional and social media websites.

Inclusion criteria
Males and females engaged in level 1 or 2 sports (self-reported), thus ≥ 50 hours/year; [32] prior to primary anterior cruciate ligament injury will be recruited. This study will involve adults aged 18 to 35 who have undergone primary, isolated and unilateral ACL-reconstruction (≥ 6 and ≤ 24 months after surgery).
Included participants need to be cleared for return to sports from one's medical team (Physician and Physiotherapist).

Exclusion criteria
Participants will be excluded if they: (I) are not regular participants in level 1or 2 sporting activities (< 50 hours/year), (II) are < 6 > 24 months after ACLR, (III) have a history of previous ACLR, (IV) have had unhappy Triad´ in the knee joint (inner ligament, inner meniscus and cruciate ligament), (V) have muscle soreness, (VI) have pain in the lower extremity, (VII) have a history of other lower extremity injuries < 6 months requiring surgical repair, (VIII) have had a concussion < 6 months, (IX) currently take analgesics regularly/any other medication(s) that might in uence the perception and physical performance, (X) have any serious illnesses that is affecting their quality of life, (XI) are pregnant or breastfeeding.

Screening and inclusion
Only participants who meet the requisite inclusion criteria will be given an appointment for the test measurement by the research team. When a participant consent to be contacted by the research team, eligibility will be determined via e-mail or telephone conversation based on the inclusion and exclusion criteria for the study. Enrolled study participants will then receive all the preliminary study documents via e-mail to enable them adequately prepare for the appointment.

Study ow and procedure
All testing measures will be performed at the Goethe University Hospital, Frankfurt (Main), Germany, Division of Preventive and Sports Medicine.
Data assessment will take place at two separate days at comparable time of day with a washout phase of one week in between the toe days. At both days, all participants will familiarize themselves with the measurements. Participants will be guided to perform six test jumps (2 anticipated and 2 each for unanticipated 1 (consistent) and 2 (inconsistent) following a standardized mini warm-up (30 jumping jacks).
After a brief break, the intervention will be carried out. Then, all participants will perform the measurements. Primary and secondary outcomes will be assessed simultaneously. At the end, some participant characterizing variables and potential confounders (e.g. questionnaires and neuromuscular performance testing) will be collected. An overview of the study ow is provided in Fig. 1.
The measurements on both days are carried out under standardised conditions (e.g. light, temperature, test order, sportswear/shoes etc.).
Preventive Warm-up protocols Two different warm ups; thus Prevent Injury and Enhance Performance (PEP) will be compared to a standard guideline Ergometer cycling of the same duration. All participants will undertake both the Prevent Injury and Enhance Performance (PEP) as well as the Ergometer cycling warm-up (see below) on two separated visits with a washout period of one week between measurements.

Randomization procedure
The order in which enrolled participants will perform the warm-up programmes will be randomized. The randomization list will determine whether a participant performs the PEP warm-up or ergometer cycling on test day 1 or 2. Microsoft Excel (O ce 2016) software was used to generate 3 different randomization lists, one each for the starting warm-up condition, starting leg for single leg hop for distance and the order of the 3 different jump conditions (anticipated, unanticipated-consistent and unanticipated-consistent) to be tested.

Prevent Injury and Enhance Performance (PEP): Intervention
The PEP programme was developed by the Santa Monica Sports Medicine Foundation's ACL injury Prevention Project group as a strategic training programme to help reduce ACL injuries in female soccer players. It is a highly speci c 15-minute (10-minutes dynamic and 5-minutes stretching) training programme targeted at neuromuscular and proprioceptive problems identi ed in previous studies [8,10,33]which have shown that neuromuscular and proprioceptive training can help reduce ACL injuries by two to four folds. This prevention program is subdivided into 5 sections (Section I: Warm-up, Section II: Strengthening, Section III: Plyometrics, Section IV: Agilities and Section V: Stretching) Since the focus of this trial is the biomechanical landing stability and decision making quality during a single leg CMJ only the dynamic component (Section I-IV): aiming to enhance neuromuscular performance and coordination of the lower extremity will be used. The warm-up programme will last for 12 minutes (including short breaks between exercises) and will be performed at a moderate intensity level (Borg Scale: [11][12][13][14]. The participants will be asked after the rst half of the programme to estimate their self-reported level of intensity. If the intensity exceeds or falls below the above-mentioned BORG range, participants will be informed to reduce or increase the tempo and the inter-exercise breaks, respectively. Immediately after the completion of the PEP programme, participants will be asked again to rate their subjective level of exhaustion before they will have a three to four minutes sitting rest (BORG ≤ 10).

Ergometer cycling: Control
The participants will cycle (cadence of 70 to 80 repetitions per minute) on an electronic Ergometer (ergoline GmbH, Bitz, Germany) at moderate intensity (Borg Scale 11-14) for 12 minutes just as the PEP intervention. The participants will be asked after two, six and eight minutes to estimate their subjective level of exhaustion. If the intensity exceeds or falls below the above-mentioned BORG range, the resistance (watt) will be adjust accordingly on an individual basis. Immediately after the completion of the cycling warm-up, participants will be asked again to rate their subjective level of exhaustion before they will have a three to four minutes sitting rest (BORG ≤ 10) just as after PEP.

Unanticipated and anticipated jump-landing assessments
Immediately after each warm-up-programme, participants will perform repetitive CMJ (with hands placed at the hip) with single leg landings on a force plate; (HUMAC Balance System, Computer Sports Medicine Inc. Stoughton, Massachusetts, USA). A calibrated HUMAC Balance System was reported by Koltermann et. al. [34] to produce high quality and biomechanical measurement outcomes with minimal errors, similar to conventional force plates under static and dynamic conditions. Three different conditions (randomly ordered) are to be completed: anticipated (n = 4), unanticipatedconsistent (n = 6) and unanticipated-inconsistent (n = 6) jump-landings (both sides equally distributed). Powerpoint slides (Microsoft Corporation, Redmond, Washington, USA) will be used to indicate the requested landing leg (left or right footprint relative to a vertical line). These slides will be shown on a 21.5-inch monitor placed at a height of 1.5 m and 2.5 m in front of the force plate. The slide depicting the requesting landing will be shown already before the jump. The take-off will lead to a release of a USB button foot switch (PC Sensor USB foot switch Keyboard; China) positioned under a plastic panel, where the participants are standing on prior to the jump. This button release will elicit a slide change with a latency of approximately 120 milliseconds (ms). Figure 2 illustrated the setup of the jump-landing task.
Regardless of the landing condition, participants will always start by standing on the plastic panel (with hands placed at the hip) while focusing the Powerpoint-slide being displayed on the screen in front of them. They are instructed to land on the side, which is illustrated before take-off (pre-de ned plan/expectation) except the landing side change during the jump. After ground contact, the landing needs to be stabilized as fast as possible. The participants are asked to start the jump after a member of the research team instructed them verbally to do so.
Under the anticipated condition (Fig. 3), participants will be informed before take-off that the second Powerpoint-slide indicating the landing foot will be a repetition of the initial slide on the screen before take-off (i.e. no change of landing side after take). Therefore, the participants will have su cient time to prepare their landing.
Under the unanticipated-consistent condition (Fig. 4), the landing side will not change during the jump as in the anticipated condition. That means that the landing information during the ight will be consistent with the landing side shown prior to the jump. In contrast to the anticipated condition, here, it is unknown to the participants during the preparation phase for the jump that the landing side will not change upon take-off. This condition will not require the participants to adapt their pre-de ned motor plans during the jump.
Under the unanticipated-inconsistent condition (Fig. 5), the landing side illustrated before the jump will change to the other side after take-off (inconsistent with pre-movement expectation) without the participants knowing this before the jump. In contrast to the unanticipated-consistent trials, this condition will require the participants to adapt their pre-de ned motor plans/expectations during the jump by switching to the other side. In both unanticipated conditions, participants will be asked to avoid speculation and guessing but to react to the stimulus appropriately.

1) Protective mat, 2) Hinge, 3) Plastic panel, 4) USB foot switch, 5) HUMAC balance force plate, 6) HDMI cable connecting laptop and USB foot switch to monitor, 7) Monitor/Screen (21.5 Inch), 8) Powerpointslide (projected) on screen prior to jump initiation (randomised)
A successful trial is de ned as holding a stable landing position for at least 10 seconds (sec.). The participants will be allowed to use their arms to equilibrate the postural sway immediately after landing. After a stable standing position has been reached, their hands need to be repositioned on the hip while focusing the footprint on the monitor in front of them. Unsuccessful trials are categorised as standing errors (touching the ground with the free leg, leaving the force plate and touching the ground with the hands, and falls) or decision errors (landing on the wrong foot, landing on both foot). Previous studies [35,36] revealed longer ight times for unanticipated jump-landings compared with anticipated landings. Therefore, participants will be encouraged to produce ight times of 400 to 500 ms, which corresponds to a CMJ height of approximately 25 to 30 cm regardless of the condition. The corresponding available response times during the jump of 300 to 400 ms ( ight times minus 120 ms latency of the button switch) are consistent with those of other studies [36,37]. In order to achieve the required ight times, participants will practice the jump-landing task (anticipated: n = 4, unanticipated-consistent/-inconsistent: n = 4 each, random ordered) during the familiarisation session at both days and will be provided with feedback regarding the achieved jump height after each jump. Those participants who are unable to reach a minimum ight time of 400 ms will be allowed to use their arms during take-off to gain momentum. If the required ight times are not reached during the actual jump-landing testing, participants will be informed accordingly and asked to consider that for the next trial. All participants will be expected to wear sports clothes (t-shirt and shorts) and indoor sports shoes during both task familiarisation session and the actual jump landing experiment. Participants will be encouraged to come along with their visual aids in case they use them during sports.

Primary outcome
The vertical peak ground reaction force (pGRF) and the time to reach this maximum value (recorded with a sample frequency of 100 Hz) upon landing will serve as the primary outcomes of this study. These values will be derived from the raw data of each trial.

Secondary outcomes
Postural landing stability: The centre of pressure (COP), time to stabilisation (TTS) and the number of standing errors will be used to operationalize the postural landing stability. Because a decreased dynamic postural stability has been found to be predictive for non-contact knee injuries in team sport athletes, the assessment of these variables is of particular relevance.
The COP path length re ects the cumulative postural sway (millimetre) of the center of mass recorded between the foot sole (area of support) and the force plate for a speci c time. [38] The cumulative deviations of the vertical ground reaction forces will be calculated for the medio-lateral and anteriorposterior directions during the rst 2.5 seconds after landing as this period corresponds to the early dynamic stabilization phase (Fransz et al. 2016). [39] The COP is considered a valid and reliable parameter for measuring postural control in different populations. [40] The TTS indicative for the time needed to regain a stable stance after landing relative to the whole stance time of ten seconds will be analysed in accordance with the speci cations provided in Colby et. al. [41] Wikstrom et. al. [42] A moderate to high reliability has been demonstrated for the TTS in Jensen et. al. [43] .

Decision making quality
In order to operationalize the quality of decision making during the jump, the number of landing/decision errors (landing on the wrong leg or both legs) will be recorded and counted. In this context, "decisionmaking quality" refers to the ability to quickly adapt motor plans/intentions and actions to an unexpected visual landing information (landing on the left or right leg) during the jump. Standing errors will be de ned as correct landings, which could not be stabilized over the ten seconds measuring period (i.e. ground contact of the free leg, leaving the force measuring plate, touching the ground/plate with the hands or fall).
The documentation of this type of error will be carried out for both the anticipated and the unanticipated conditions, although, this kind of error should only occur under the inconsistent unanticipated trials, because the other conditions do not require task switching. A high decision-making quality results in a lower number of decision errors and vice versa.

Potential confounders
The following potential confounding variables will be assessed: Single leg hop for distance (SLHD; maximum value of each side [cm] from three attempts). The SLHD is a valid and reliable test for assessing leg strength, neuromuscular control and dynamic knee stability in individuals after ACL reconstruction. [44] The reliability of this test is high with ICC of 0.97 (CI 0.9-0.99) and a standard error measurement of 3.5%. [45] A limb symmetry of 90 percent (jumping distance) is considered an essential prerequisite for the resumption of sports and competition activity after ACL reconstruction [44,46,47] Self-reported knee function of the ACL-reconstructed and uninjured knee using the Lysholm knee scale (0 = maximum restricted knee function; 100 = no restricted knee function at all; Lysholm & Gillquist, 1982; all test subjects). Good psychometric quality was demonstrated for the questionnaire. [48] Fear of re-injury associated with anticipated and unanticipated jump landings assessed by 10 cm visual analogue scale (0 = no fear at all, 10 = maximum imaginable fear; all test persons) during the familiarisation session.
Mean ight time differential within and between anticipated and unanticipated jump-landing conditions (ms).
Ability in selected cognitive functions (working memory and mental exibility) found to be relevant to make fast and accurate landing-related decisions under time-constraints assessed via the trailmaking test version A and B [49] (time to test completion in seconds; higher values indicating lower test performance) after the jump-landing tasks on both days.
Participants' current level of daily physical activity will also be evaluated using the the international physical activity questionnaire short form (IPAQ-SF). The validity (predictive, concurrent, convergent, criterion and discriminant) of this research tool is detailed in Lee et. al., [50] with a high reliability of α < 0.80. This questionnaire will be used to collect information from participants about time spent being physically active in the last 7 days.

Statistics
Parametric or non-parametric statistical tests will be selected after checking for all relevant underlying assumprions (data distribution, variance homogeneity and linearity).
All data will be reported descriptively as mean values, minima and maxima, standard deviations and 95% con dence interval or, in the case of non-parametric analyses, in form of median and interquartile ranges.
The statistical analysis will be based on the intraindividually mean values/medians of the primary and secondary outcome measures. All outcome measures will rst be tested for potential differences between the ACL-reconstructed and uninjured side.
Potential effects of both warm-up protocols will be evaluated for both sides separately (including limb symmetry) and non-side-speci c (both limbs together).
To identify potential carry-over effects between the two interventions, unpaired t-tests will be calculated to compare the intraindividually sums of both sequence groups (starting with PEP or cycling ergometer warm-up).
To investigate the implications of the assessed potential confounders on the between-condition differences explorative correlation analyses will be used. Potential signi cant between-condition ight time differences will be considered as covariates in the statistical analyses.
For the between-legs and between-warm-up programs' analyses, repeated measures analyses of variances ANOVAs (with the potential confounders as co-variates) or Friedman tests will be calculated.
In case of signi cant differences, Bonferroni-Holm adjusted post-hoc testing (paired t-tests/Wilcoxon tests) for the individual within and between-condition differences (unanticipated landing costs) will be applied.
The signi cance level (alpha error) will be set at 5%t. All statistical analyses will be performed using SPSS (

Discussion
The ndings of the planned trial will help clinicians, therapists and athletic trainers to better estimate the suitability of a team-sport speci c injury preventive warm-up programme to enhance acutely the ability of ACL-reconstructed athletes to deal with unanticipated circumstances on the pitch.
One major strength of our study is the fact that we will measure jump landing performance under unanticipated instead of just unplanned conditions as done in most of the related studies. This may better re ect the real playing situations where athletes do not know initially, if the circumstances will allow them to realize a pre-de ned motor-plan (consistent with pre-movement expectation) or require them to adapt their motor-plans (inconsistent with pre-movement expectation) during a jump. Furthermore, this distinction will enable us to identify those, who tend to guess or to just realize their initial expectations regarding the landing side without reacting to the stimulus during the jump (i.e. decision errors during consistent unanticipated trials), which was a limitation of a previous trial of our working group (Giesche et al. 2019). [35] Another strength relates to the fact, that we only considered physically active participants with a postoperative time within two years thereby increasing the homogeneity of our sample.
To sum up, the knowledge gained by our planned trial may provide novel insights into the implications of previous warm-up routines for potential secondary ACL-injury risk factors occurring under challenging athletic affordances.

Trial status
The data collection for this study started in June, 2019 and the rst participant was assessed on June 25, 2019. Recruitment is still ongoing at the time of submission of this manuscript. Final enrolment is planned to be completed by the end of August, 2020 and statistical analysis of the assessed data is expected to be completed by October, 2020. This is the rst version of the protocol which was registered in the DRKS on May 24, 2019.

Declarations
Availability of data and materials: Not applicable.
Ethics approval and consent to participate: The study was approved from the local Ethics Committee of the Faculty of Psychology and Sport Science, Goethe-University Frankfurt (reference number: 2017/27). A written informed consent will be obtained from each participant. The data obtained will be anonymously kept and used only for the purspose of the study.

Not applicable
Competing interests:

None declared
Funding: This study received no grant from any funding agency in the public, commercial or not-for-pro t sectors.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.