Masters Research Study: The effect of zero drop shoes on measures of running economy in ultra-trail runners. — Pacer

Running is a form of locomotion which is differentiated from walking by the presence of a  flight phase which replaces the double stance phases of walking (Lohman, Balan Sackiriyas and Swen, 2011). The four phases of running can, therefore, be classified as the stance, early swing or float, middle swing, and late swing or float phases (Pink et al., 1994; Lohman, Balan Sackiriyas and Swen, 2011). Furthermore, the initial ground contact of runners can be described as: rearfoot strike (RFS), where the calcaneus contacts the ground first, midfoot strike (MFS), in which the rearfoot and forefoot meet the ground simultaneously, and forefoot strike (FFS), where the forefoot lands on the ground first followed by the heel (Lieberman et al., 2010). Running related injury (RRI) incidence is reported to be high. The Runners and Injury Longitudinal Study (TRAILS), conducted on 300 initially uninjured runners during a 2-year observation period, observed that at least one overuse running injury was sustained by 66% runners, with 56% of the injured runners being injured more than once within the 24-month observation period (Messier et al., 2018). Runners have sought ways to reduce injury rate, with one approach being the shift to more minimalist running shoes. There has been an ongoing debate about whether foot strike patterns play a role in running-related injuries (RRIs). In an acute phase, changing shoes from regular to minimalist has been found to shift runners from a predominantly rear-foot striking (RFS) pattern to more of a forefoot (FFS) pattern (Lieberman et al., 2010; Larson et al., 2011; Daoud et al., 2012).  One of the arguments for a longitudinal shift from RFS to FFS has been largely driven by anecdotal evidence that running in minimalist, ‘natural’, footwear is related with a FFS pattern, and therefore running in them could reduce running related injuries (Martin Lou et al., 1985; Warne and Gruber, 2017).  Both FFS and RFS runners show high injury rates, however, RFS runners are 2.6 times more likely to have mild injuries and 2.4 times more likely to have moderate injuries, and a nearly two-fold severe injury rate when compared to FFS runners (Daoud et al., 2012). Research into the effect of zero-drop, cushioned shoes on foot strike pattern is sparse, with one study demonstrating a highly varied result when habitually RFS, regular-drop shoe runners changed to either minimalist zero-drop, traditional zero-drop, or maximalist zero-drop shoes after a four-week training period (Deneweth et al., 2015). Deneweth et al. (2015) observed that one runner transitioned to a FFS style after a four-week retraining period in minimalist shoes, while the other three subjects maintained a RFS style, with either an increased or decreased peak loading rate (Deneweth et al., 2015).Foot strike pattern plays a significant role in the lower extremity mechanics during the early stance phase of running (Valenzuela et al., 2014; Almeida, Davis and Lopes, 2015), in particular, by modulating impact forces associated with running. Impact forces are hypothesized to contribute to some RRIs because they generate a shock wave that travels up the body, generating potentially high stresses and strains on musculo-skeletal tissue which can eventually lead to injury due to the cyclical nature of running. Runners who use a natural RFS pattern exhibit significantly higher vertical loading rates (VLRs) compared to natural FFS (Almeida, Davis and Lopes, 2015), and greater VLRs have been linked to running injury, particularly tibial stress fractures (Zadpoor and Nikooyan, 2011) and plantar fasciitis (Pohl, Hamill and Davis, 2009). It is further noted that running using a FFS or MFS pattern displays a lower impact peak ground reaction force (GRF), described as a marked, substantial impact peak that is superimposed on the upslope of the vertical GRF immediately after the foot’s initial contact with the ground, when compared to RFS (Lieberman et al., 2010). These higher rates and magnitudes of impact loading have been shown by some studies to correlate significantly among RFS runners with lower limb stress fractures (Milner et al., 2006), plantar fasciitis (Pohl, Hamill, & Davis., 2009), and other injuries such as hip pain, knee pain, lower back pain, medial tibial stress syndrome, and patellofemoral pain (Milner et al., 2006; Hamill et al., 2008).The relationship between RFS, regular running shoes, and RRI has arguably contributed greatly to the shift of runners from regular shoes to minimalist shoes. Previous studies have shown that trained regular-drop runners modify their running technique during an acute bout of barefoot running and exhibit a less dorsiflexed ankle at initial contact with a larger ankle range of flexion during the stance phase and a more flexed knee joint with a lower knee joint range of flexion during the stance phase (Chambon et al., 2015), which is shown to lead to a decrease of the first peak of vertical GRF (transient peak) (Giandolini et al., 2016). This indicates that minimalist shoes should provide the platform for more of these positive gait changes due to the fact that they are more similar to barefoot running demands and, therefore, could lead to a decreased injury risk (Daoud et al, 2012 ,Lieberman et al., 2010; Michael, Richard and Lee, 2018). Shoe drop height is the difference in stack height between the fore and rear parts of the sole of the shoe (Figure 1). While not all zero-drop shoes can be defined as ‘minimalist’ (some zero drop shoes can still have regular or even very thick soles), all minimalist shoes are zero-dropped. As such, a zero-drop shoe might result in certain kinetic and kinematic changes similar to those seen in minimalist running studies in comparison to regular drop shoes.While the reduction in impact force variables and potential reduction in associated RRI rates are one reason why runners might consider switching from a habitually RFS pattern to MFS or FFS, another major motivating factor is the potential improvement in running economy (Divert et al., 2005; Franz, Wierzbinski and Kram, 2012). Several studies have examined foot strike, with some observing that top finishers of short, middle, and long distance events tended to use an MFS or FFS (Hasegawa, Yamauchi and Kraemer, 2007; Larson et al., 2011; de Almeida et al., 2015). Running economy (RE), which is the reflection of the amount of oxygen (O2) required to maintain a given velocity, is thought to be a key factor for the determination of running performance (Sinclair et al., 2013). There are many ways shown to improve RE, with one such factor being altering lower limb kinematics. Studies comparing economy and kinematics of running in different footwear conditions conventionally derive from treadmill or over-ground experimentations using flat or level terrain (Divert et al., 2008; Franz et al., 2012; Perl et al., 2012). However, trail running is an athletic activity consisting of running outdoors on surfaces such as gravel roads, mountain hiking trails, beach trails, and mountain bicycle single track and involves regular bouts of uphill running and/or walking (Boudreau and Giorgi, 2010). There has been a large growth of the popularity of trail running in recent years, with one reason possibly being due to that participants report benefits such as increased self-efficacy, more mental focus on career goals, a more positive attitude to life, enhanced work performance, increased problem solving, improved time-management, and better organisational skills (Boudreau and Giorgi, 2010). Further, the low compliance of running on roads, such as asphalt and concrete surfaces, demonstrates higher impact forces when comparing to gravel or grass surfaces (Dolenec, Štirn and Strojnik, 2015), which can predispose runners to more overuse injuries (Rolf, 1995). There is also an increased risk of running-related injury due to the cambered nature of roads (OConnor and Hamill, 2002). Runners who run predominantly on road also have a high occurrence of injuries, up to 79% incidence, with a large percentage of this being of the knee (Taunton et al., 2003; Van Gent et al., 2007). However, Malliaropoulas et al. (2015) reported that at least one running-related injury was reported by 90% of a sample of 40 ultra-trail runners (36 men and four women), although running in the mountains (p=0.0004) was found to be a protective factor. The combination of injury risk in road runners as well as the potential protective benefit of trail running has arguably contributed as one of the main reasons why there has been a growth in trail running in recent years, and in particular, ultra-trail running (David, 2010).While research into the biomechanical aspects of running in trail runners has expanded in recent years (Björklund, 2019; Khasseterash, 2019; Vercruyssn, 2015; Vernillo, 2019), the literature in ultra-trail running is minimal. There has been further growth in the popularity of ultra-trail runners (David, 2013), with a major contributing factor in the growing popularity of ultra-trail running arising with the publication of two books: Ultramarathon Man: Confessions of an All-Night Runner (2005) by Dean Karnazes, and Born to Run (2009) by Chrostopher McDougall (David, 2013). To highlight this growth, In 1996, there were 17 trail 100-milers in the United States. In 2008, there were 59, (Graubins 2008). There were a total of 32,352 161km finishers from 1977 to 2008, with the numbers of finishers as well as events exponentially increased during that time. (Hoffman, Ong and Wang 2012), which has only increased since 2008. Ultrarunning Magazine, which chronicles and publishes race results, has indicated that “by the end of 2009, Ultrarunning had reached 36,106 individual finishers. By the end of 2010, over 46,280 individuals had reportedly finished an ultra marathon” (Lacroix 2012). However, even though there has been substantial growth in the sport of ultra-running, research in the topic is very sparse. A search in the online database, ‘Dimensions’, reveals 1159 publications since 2005, with only 23 using the term ‘ultra-trail’ in the title and abstract. This is compared to 3009 articles citing ‘trail-running’ for the same period. Overall, an improved running economy has been closely associated with improved performance in highly trained athletes (Conley and Krahenbuhl, 1980; Hoogkamer et al., 2016; Lucia et al., 2006; Williams and Cavanagh, 1987) as well as shown to be beneficial for reducing injuries (Daoud et al 2012). The literature suggests that running in minimalist shoes may be beneficial for improving running economy in road running, but there is minimal literature examining this effect in trail running, specifically ultra-trail running, nor in cushioned zero-drop shoes. As such, the purpose of this study is to determine whether there are differences in kinematic and gait spatio-temporal factors between zero and regular drop ultra-trail marathon runners, and whether these factors are related to running performance.Research Aim OneThe first aim of the study will be to compare the acute differences in kinematic, and gait spatio-temporal factors in regular-drop and zero-drop runners, during three bouts of submaximal flat treadmill running in recreational ultra-trail runners.ObjectivesThe objectives that will guide the researcher for aim one will be to compare the following variables between zero drop and regular drop recreational ultra-trail runners during flat submaximal treadmill running at three speeds:the peak knee angle at ground impact measured with the use of the Xsens Motion Analysis Technology.angle of plantarflexion measured with the use of the Xsens Motion Analysis Technology.foot striking pattern measured with the use of the Optagait system.the average vertical oscillation with the use of the Optagait system.trunk forward lean measured with the use of the Optagait system.stride length and frequency measured with the use of the Optagait system.HypothesisIt is hypothesized that the zero-drop recreational ultra-trail runners will differ from regular-drop ultra-trail recreational runners by:having a greater angle of plantarflexion and a smaller knee angle at ground contact, displaying a greater distribution of mid-or fore-foot strike, having a lower vertical oscillation and a smaller angle of trunk forward lean in relation to the running surface throughout the gait cycle,having an increased stride frequency, andhaving a decreased stride length.Motivation and Potential BenefitsThis study will provide valuable information with regards to running biomechanics for recreational ultra-trail runners running in both regular-drop shoes and zero drop shoes. By examining the influence of shoe-drop height on RE in ultra trail runners, the study will aid the current body of literature regarding trail running and performance. Furthermore, it could provide information to shoe manufacturers by providing evidence for their claims, for both development and marketing purposes, as well as provide valuable information for individuals who wish to convert from regular-drop shoes to zero drop. Because of the increasing popularity in ultra-trail races, any differences in the economy related variables (ground force variables, joint kinematics, and fitness parameters) could help with further motivation for the  prescription of shoes to individuals who wish to convert from running in  regular-drop shoes to running in zero-drop shoes. MethodologyStudy DesignThis study will be descriptive and observational, with no intervention. Similar to the protocol utilised by Folland et al. (2017), all tests will be completed in the morning (07:30–12:00) at a laboratory temperature of 18–20°C. Participants will be instructed to arrive at the laboratory well hydrated, having avoided strenuous activity for 36 hours, alcohol for 24 hours, and caffeine ingestion for 6 hours before testing. Prior to the test, participants will be informed and familiarised with the testing procedure, and given the necessary amount of time to warm up effectively. This study will involve two groups of ultra-trail runners, ~15 who habitually run in zero drop and ~15 who habitually run in regular-drop shoes. Regular-drop shoes will be defined as shoes with a heel-to-toe drop of between 6 and 10 mm. Participants will run on a treadmill at level/flat at three speeds predetermined via a VO2 max test. VO2 data will be measured by breath-by-breath analysis. Kinematic and spatio-temporal variables will be measured by the xsense motion analysis technology ( and 2D video analysisIn order to be included in the study, participants are required to have been habitually running in their shoe-type (regular or zero-drop) for at least 95% of their training and racing in the past 12 months, run at least 50% of their training runs on trail, be injury free for 3 months (Schütte et al., 2016), and be between the ages of 19-55. Participants would need to have run one or more ultra-trail races (>42.2km) in the past 6 months. All trail runners will wear their own habitual running shoes, and all shoes should be in good conditions of use and have similar characteristics of construction (Da Silva Azevedo et al., 2016).This study will consist of two parts, separated by one week. On the first day of testing (part one), VO2 max will be collected at the Stellenbosch University Sport Physiology Laboratory ( to determine peak treadmill speed for the calculation of the three submaximal speeds. Data will be collected through breath-by-breath analysis (COSMED), specifically VO2, Respiratory Quotient (RQ), and anaerobic threshold (AT). The VO2 Max protocol used will be the Vameval protocol. Participants will start at 11km/h for men and 9km/h for women at an Incline of 1% for the entire test. Speed will increase with 0.5km/h every minute until exhaustion. Prior to treadmill running, anthropometrics will be measured and recorded for each participant, including body mass, height, thigh length and circumference, calf length and circumference, foot length and breadth, and malleoli height and width (Boyer 2015). The participant will be given a series of questions for them to complete relating to their running history and injury prevalence.On the second day of testing (part two) the gait spatio-temporal and kinematic data will be collected during submaximal treadmill running at three predetermined speeds (6 minutes at 60 % and at 70 % and 3 min at 85 % of their personal peak treadmill running speed.). The skeletal joint kinematics, such as angle of trunk lean, vertical oscillation, peak knee angle, knee flexion during stance, delta knee angle, peak hip angle, and delta hip angle will be monitored with the use of the Xsens Motion Analysis Technology, and spatio-temporal variables, such as stride angle, ground contact time, swing time and stride length and frequency will be measured for every step using an optical measurement system (Optagait, Microgate, Bolzano, Italy) placed at treadmill belt level, will be measured  during the same submaximal running trials. The order of part one and two of testing will be completely randomised for each participant and completed within one week of one another, but not within 48 hours, thereby eliminating any potential learning effect between the two trials.Proof of ConceptA proof of concept study will be conducted by the primary researcher (Mr R Henning) and supervisor (Mr S de Waal) to evaluate if the proposed methodologies are possible. The primary researcher is a habitually zero-drop runner, while the supervisor is a habitually regular-drop runner, and they should hold the same inclusion criteria to the main study. Duration of the StudyThe study will begin in February 2021, with the write-up of the research protocol and study design. Whilst awaiting ethical approval, a proof of concept study will be conducted on a sample of two runners, between May and June of 2021. After ethical clearance has been granted, marketing and participant recruitment will begin in June 2021 until July 2021. Testing and data collection will begin in August of 2021 and end in September 2021. The data will then be analysed during October and November of 2021, followed by the final write-up and hand in date of August 2022.Participants and SamplingA sample of convenience will be used, whereby runners from around the Cape Town area will be recruited through social media and personal networks. The study will follow a similar inclusion/exclusion criteria to Kasmer et al, (2014), and will require 15 recreational trail runners per group (i.e. 15 zero drop and 15 regular drop runners) for a total of 24 – 30 participants. Participants can be either male (18-45 years old) or female (18-55 years old). Regular drop and zero drop groups will be matched for distribution of sex (i.e. if 60% of regular drop participants are female, then ~ 60% of zero drop participants will also be female).An information advertisement (Appendix B) will be sent out to Facebook running groups in Cape Town (Cape Town Runners, Cape Town Trail Runners) as well as to internal networks (Altra Running ZA Ambassador and Athlete group, organisers of trail runs: Ultra-Trail Cape Town™, Maxi-Race™ Cape Winelands, Addo Elephant Trail Run, Chokka Trail Run) including a summarised version of the study inclusion and exclusion criteria. Contact details of the researcher will be provided on the advertisement and individuals who are interested can contact the researcher directly. Individuals who fall within the relevant inclusion criteria will be contacted to ask to participate in the study. To be included in this study, the individual needs to be between the ages of 19 and 55 (Kasmer 2014). Each participant will meet the American College of Sports Medicine (ACSM) guidelines for exercise testing and prescriptionfor physical function based on questions from participant’s pre-protocol questionnaire (Appendix C) and should be asymptomatic for cardiovascular/pulmonary disease. Runners in the zero-drop sample should hold at least one year experience primarily running in zero-drop shoes, run greater than 42.2-km in zero-drop shoes within the past six months and run greater than 64.4 km (40 miles) per week. They should have; no injuries within the past three months as defined by medical treatment or stoppage of training for greater than one week due to injury, no current injury, (Schutte 2016) and have the ability to follow the study protocol.EthicsApplication for ethical clearance was approved by the  Human Research Ethics Committee of Stellenbosch University and the Research Ethics Committee, on 1 July 2021, where after participants for the study will be recruited. An oral presentation of this protocol will be conducted in front of the Department of Sport Science, FMHS, SU on 11 or 18 May 2021 with written feedback provided on changes required prior to HREC submission. All information, data and any other personal details will only be accessible to the researcher, supervisor and co-supervisor. All information will be stored in two different manners. Firstly, all documents will be safely stored on the researcher’s personal password protected computer and will be automatically backed-up onto the researchers personal google drive cloud services and  will be backed-up onto a flash drive on a weekly basis and given to the supervisor for revision and safe keeping. Secondly, all signed documents (informed consent, pre-screening questionnaire, COVID screening) will be stored in the primary supervisor’s office (Mr S. de Waal, room 409, Department of Sport Science, Stelenbosch University) , in a cupboard which will be locked at all time. Only Mr de Waal will have the key, and access will only be granted under his supervision. Documents will be stored for five (5) years maximum. COVID-19 ProtocolBoth Stellenbosch University ( and Government gazetted ( COVID-19 protocols will be adhered to at all times during the testing process. However, the following measures will be adhered to as standard practice throughout the testing process:The laboratory testing (part one and two) will be kept to a maximum of four persons in the lab space at all times including; one participant only, the primary researcher and study supervisor, and one lab technician.The researcher, supervisor, and lab technician(s) will keep their masks on at all times throughout the testing procedure. The participant may only take his/her mask off at the onset of the testing procedure and must put it back on once testing is complete.Social distancing will be maintained as much as possible, with the only times this will not be possible is when taking anthropometric measurements, and attaching the IMU or COSMED devices. Mask wearing will be adhered to by both parties when taking anthropometric measurements and attaching the IMUs, and the participant only will be allowed to remove his/her mask when being fitted for the COSMED.Both the participant and researcher, supervisor, and technician(s) will regularly sanitize their hands using alcohol based hand-sanitizer, specifically when; entering the laboratory space, after fitting any equipment to the participant, after touching any apparatus (including signing of informed consent forms for example), and when leaving the laboratory area.Completing a COVID-19 screening tool (the participant must complete a manual sign in at the entrance to the Department of Sport Science, while the other parties will complete the PowerApps screening survey) on the day of testing(s).Anthropometric assessmentsPrior to the VO2 max testing, certain anthropometric measurements will be taken. These will include height (cm), weight (kg) and circumferences (cm) of the calf and middle thigh. The testing will be conducted by the primary researcher and documented on the participant’s personal Excel file.VO2 max Testing VO2 max testing will be conducted by the Stellenbosch university Sport Physiology Laboratory to determine RE via breath-by-breath analysis. VO2 max will be collected at the Stellenbosch University Sport Physiology Laboratory ( to determine peak treadmill speed for the calculation of the three submaximal speeds. Data will be collected through breath-by-breath analysis (COSMED), specifically VO2, Respiratory Quotient (RQ), and anaerobic threshold (AT). The VO2 Max protocol used will be the Vameval protocol. Participants will start at 11km/h for men and 9km/h for women at an Incline of 1% for the entire test. Speed will increase with 0.5km/h every minute till exhaustion. Participants will be instructed to wear their usual running attire for a moderately temperate climate, as well as their usual trail running shoes. Upon arrival in the laboratory, participants will complete the informed consent form, COVID-screening questionnaire, as well as the PAR-Q form, followed by their Anthropometric assessments as described above. The primary researcher will then describe the protocol to the subject, followed by the placement of the chest strap heart rate monitor (Garmin Ltd., Germany) as shown in figure 2. Participants will then step on to the stationery treadmill and commence with a 5-minute treadmill warm up at 8 km/hour. Following this, the treadmill will be stopped and the subject remains on the treadmill while wearing the safety harness and face mask for the breath-by-breath analysis. The protocol, which is loaded on to the COSMED system, will begin upon indication by the subject and the treadmill will increase in speed as according to the protocol. The protocol will stop as soon as indicated by the test subject by means of them placing their hand on the front railing on the treadmill, or until complete exhaustion. VO2 max is then determined by using the last 30 seconds of the highest workload during testing.Kinematic, and spatio-temporal testingRandom allocation of participants during the running test will be determined by a random/rank table that will arbitrarily assign a testing order for each participant (Refer to Appendix C). Participants will be required to run on a treadmill at level a for 6 minutes at 60 % and at 70 % and 3 min at 85 % of their personal peak treadmill running speed determined by the VO2 max testing during day 1. Participants will be instructed to wear the exact same clothing as for the original testing. Upon arrival at the Sport Physiology Laboratory, participants will be briefed on the protocol and fitted with the Xsens motion capturing accelerometers on the following locations as shown in figure 3: the posterior skull (slightly above the occipital protuberance), mid upper arms (L + R), wrists (L + R), mid-upper sternum, along the upper scapula (L + R), mid upper leg (L + R), proximal tibia (L + R), and upper foot (L + R). The accelerometers are fitted either into sleeves on a tight-fitting upper body suit, or by non-adhesive material on to straps on specific locations. Data from the Xsens motion capturing accelerometers will be captured and interpreted by the MVN analyze software installed on the principal investigators personal laptop. The Optogait will be set up on either side of the treadmill belt, along with 2 cameras set up directly in front and at a 90 degree angle alongside the treadmill. Participants will further be fitted with the COSMED face mask. Participants will step onto the treadmill and conduct a warm up at 8 km/hour for 5 minutes immediately followed by the testing protocol of 6 minutes at 60% and 70%, and 3 minutes at 80% of peak VO2 max speeds determined by testing day 1. There will be no rest between the warm up and each designated speed, and as each test subject's speed will likely differ, the treadmill will be operated manually by the primary researcher.
© Davy Mellado, 2022. All rights reserved.