The Effect of Basketball Shoe Collar on Ankle Stability: A Systematic Review and MetaAnalysis

Authors: {'first_name': 'Ci', 'last_name': 'Jiang'}


Purpose: This review is to systematically estimate if there is any relationship between the shoe collar characteristic and the risk of ankle sprain by reviewing the literature.

Methods: Relevant papers were collected through PubMed, ScienceDirect, Google Scholar, and Scopus focused on ankle stability comparison in basketball shoes between high-top and low-top. Articles were involved if they: 1) compared subjects between high-top shoes and low-top shoes, 2) only focused on basketball shoes, 3) utilized kinematics and electromyogram as the primary outcome, 4) published in English.

Results: 6 studies were collected in this systematic review, the selecting studies were divided into two different groups based on different motions of test, one of group is a side step cutting task for testing ROM of ankle in coronal plane, another one is conducting a jump task and the peak ankle inversion data at the landing after jump was extracted. Through the meta-analysis combined with a random effect, the model indicated that the type of shoe has no effect on the ankle inversion (p = 0.08) and ROM (p = 0.85).

Conclusion: The results showed that there no significant difference in the shoe collar was found. Thus, the more parameters of ankle changes and muscle activity of lower extremity should be assessed in the future to improve our comprehensive understanding for shoe collar effect of ankle stability.

Keywords: basketball shoeshigh-lop shoeslow-top shoesshoe collarankle sprainshoe design characteristic 
 Accepted on 22 Dec 2019            Submitted on 09 Dec 2019

1. Introduction

There appears to have a greater risk for obtaining an ankle sprain in basketball when compared to other sports, with incidence rates of 3.85 per 1000 player. And more than 30% of the participators suffered from ankle sprains with chronic ankle instability (Ryan 1994). Then subsequent sprains are significantly increased due to the initial experience of an ankle sprain. According to Smith and Reischl finding that 80% players who experience an ankle sprain have sustained multiple sprains (Smith and Reischl 1986). It emphasizes the potential severity of ankle sprains on the movement performance in the longer-term. These lateral ligament sprains in ankle usually appeared in the phase of touchdown accompanied by excessive inversion and plantarflexion of the foot when players land on an uneven or unbalanced surface with cutting maneuvers or changing direction (Wright et al. 2000). It is important to undertake some interventions to help prevent ankle sprains based on the prevalence of these injuries as well as the residual symptoms. Such as heel cord stretching, peroneal and tibial muscle strengthening, proprioception enhancement training, ankle braces and shoe selection have been used to reduce injury risks of ankle sprain (Knight and Weimar 2011). The ankle joint can be theoretically supported by external equipment such as ankle braces, high-top shoes and taping which can limit the range of motion of ankle joint to decrease risk of injuries, and the high-top shoes are designed as an alternative of taping and ankle bracing. The possible biomechanical reasons for those devices have mainly limited the range of movement (ROM) of ankle inversion and reduce the stress in external joint. Actually, the impact of high-top shoes in inhibiting ankle sprain has been widely researched since the 1980s. However, there is no scientific consensus on the stabilizing effect of high-top shoes in limiting ankle inversion (Blache et al. 2011).

There are a number of studies reported that compared with low-top shoes, the high-top shoes reduce the inversion angle of the ankle joint (Ottaviani et al. 1995; Ricard, Schulties, and Saret 2000). Not all studies have shown that high-top basketball shoes are more advantageous than low-top basketball shoes in preventing ankle sprains. In contrast, there are no significant differences were presented in ankle sprain between high-top shoes and low-top shoes for those who had inexperience of ankle sprains (Barrett et al. 1993; Handoll et al. 2001). Furthermore, the prospective, randomized experimental was designed to estimate 622 subjects for duration of the season to observe if there are any difference between the high-top shoes and low-top shoes by Barret et al., and the result showed that there no significant differences were found smaller risk of ankle sprains while taking high-top shoes (Barrett et al. 1993). Similarly, high-top shoes did not emerge superiority characteristics over the low-top shoes in reduce risk of ankle sprain, especially for those who had no history of ankle sprain (Rovere et al. 1988). Therefore, there are conflicting results in the literature refer to effectiveness and superiority stability in high-top basketball shoes.

Meanwhile, the pre-landing muscle activity of lower extremity such as peroneus longus, peroneus bervis, and tibialis anterior exhibit strongly influence muscle stiffness which facilitates the elastic energy stored and released, also limits the ankle to be excessive inversion and plantarflexion after foot landing (Mrdakovic et al. 2008; Neptune et al. 1999). According to the researcher indicated that due to the various shoe designs which include cushion properties and ankle collar features may provide subjects with different neuromuscular responses of lower extremity muscle stiffness at the touch-down (Neptune et al. 1999). Based on those, some reports indicated that the difference was found in lower leg muscle activity between the hip-top and low- top shoes. There is a piece of evidence that shoe features can substantially influence muscle contraction during the sudden inversion of ankle (Kerr et al. 2009; Ramanathan et al. 2011). However, there are only few studies to demonstrate the relationship between the basketball shoe characteristic (high-top and lower-top shoes) and ankle sprain from the perspective of muscle activity.

The purpose of this review is to systematically estimate if there is any relationship between the shoe collar characteristic and the risk of ankle sprain by reviewing the literature. We will provide evidence from the aspect of the kinematic and muscle activity of lower extremity.

2. Methods

This systematic review was conducted according to previously recommended guidelines and was followed in accordance with the PRISMS statement (Liberati et al. 2009).

2.1. Search Strategy

The system database retrieval was executed by PubMed online as the main database. What more, ScienceDirect, Google Scholar, and Scopus were carried out as complementary databases, the relevant studies also were extracted in the reference lists of published. The detailed electronic database search strategy was as follows: basketball shoes, high-top shoes, low-top shoes, ankle collar shoes, ankle sprain, ankle inversion, landing, muscle pre-activity, shoe design characteristics.

2.2. Inclusion Criteria

This review included an experiment comparing the effects of basketball high-lop shoes for ankle stability versus lower-lop shoes. We used a definition of high-lop shoes and lower lop shoes as presented in the market (Figure 1). And the articles should be meet the following inclusion criteria:

  1. The subjects were not allowed to have any experience of ankle sprain or any injuries in the lower extremity.
  2. Primary results should include kinematics parameters.
  3. The purpose of studies only focused on basketball shoes in a male player.
  4. The paper should be the English language, and the conference articles were excluded.
Figure 1 

High-top shoe and low-top shoe.

2.3. Study selection

We selected paper by reviewing all titles and abstracts for collecting publications. The full text was extracted in order to further judge whether the paper was eligible for inclusion criteria if browse the abstract is not sufficient.

2.4. Data Extraction

The data was extracted from the eligible studies which including inclusion criteria, the number of subjects was included, detailed experimental protocols, and a summary of findings. Kinematic data were drawn from six studies that involved angle of ankle inversion, range of movement of ankle coronal plane. The collection of these variables was based on repeatability and validity of extraction from the final selection studies. Then all these data are included in the meta-analysis.

2.5. Statistical analyses

Forest plots with random-effects models were used to estimate and compare the kinematic variables between the high-top shoe group and lower-lop shoe group. The random-effects model in most cases is more stable than fixed-effect model due to the weight is trivialized by sample size (Borenstein, Hedges, and Rothstein 2007). In addition, standardized mean differences, 95% confidence intervals, and p-values were calculated. Heterogeneity was assessed using I2 which was divided into different levels: high (>75%), moderate (moderate (50%–75%), and low (25%–50%) (Higgins et al. 2003). All analyses were conducted by RevMan (Version 5.3, The Cochrane Collaboration, 2014).

3. Results

3.1. Search results

900 articles were selected through four electronic databases: Scopus, PubMed, Web of science. 3 studies were extracted by a citation analysis of relevant papers. After deleting duplicates, 567 articles remained by screening titles and abstracts of collection paper. Then 500 articles were removed since they did not match the inclusion criteria. We reviewed 70 papers in full text and removed 64 because they did not eligible for the inclusion criteria. Eventually, 6 articles were involved in this systematic review and conducted the meta-analysis (see Figure 2).

Figure 2 

Flowchart of the study search and inclusion.

3.2. Basic characteristics of included studies

The risk bias was presented at Figure 3, and Table 1 showed the reasons consideration for these biases. The Kappa value was 0.788 conducted by SPSS 19.0 (SPSS, Inc., Chicago, IL, USA), it was indicated a valid for collection articles.

Figure 3 

Risk of bias of collecting articles. +, no bias; –, bias; ?, bias unknown.

Table 1

Summary of risk of bias.

Risk of bias Number of RCTs and Why

Selection bias Six RCTs did not present the method of randomization
Performance bias One RCTs instructions subjects landing on uneven plate
Detection bias Two RCTs did not use three-dimensional kinematic analysis to collect the data, may led to a deviation for the data
Report bias Three RCTs did not present outcome of all measured data
Other bias Three RCTs did not have enough sample size

Note: RCTs, randomized controlled trial.

3.3. Meta-analysis

For the meta-analyses, we concentrated on alterations in the kinematics of the ankle joint angle in the coronal plane as well as a range of motion, although many of the kinematic parameters are included in the collected articles. For example, Brizuela et al. presented the data details of kinetic variables and ankle dorsiflexion and eversion angle (Brizuela et al. 1997). Fu et al. demonstrated the maximum ankle inversion, angular velocity of ankle inversion, and time to the ankle inversion angular velocity (Fu et al. 2014). Liu et al. indicated the ankle stability from ankle peak inversion angle, inversion velocity of ankle, and total range of inversion of ankle angle (Liu, Wu, and Lam 2017). Mark et al. noted the amount of inversion of ankle as well as average rate of inversion (Ricard, Schulties, and Saret 2000). Gilbert et al. and Andrew et al. described the kinetics variables and total range of movement in ankle (Greene et al. 2014; Lam et al. 2015). According to these, there only two parameters were extracted to conduct the meta-analysis, which included maximum inversion angle and range of movement in ankle due to these two values are discussed more often than other parameters.

3.3.1. Ankle inversion and range of movement

The meta-analysis combined with a random effect model indicated that the type of shoe has no effect on the ankle inversion and ROM (Figures 4 and 5). The selecting studies were divided into two different groups based on different motions of a test, one of group is a side step cutting task for testing ROM of ankle in coronal plane, another one is conducting a jump task and the peak ankle inversion data at the landing after jump was extracted. Then the peak ankle inversion three studies were involved in the meta-analysis, which presented an overall effect p = 0.08. In the selecting articles, there two articles were presenting a significant difference in ankle inversion between the high-lop shoes and lower-lop shoes, on the contrary this condition was not found in another one article. Then, the meta-analysis presented that the overall effect on ankle range of motion did not show any difference between the high-lop shoes and lower-lop shoes (p = 0.85). In fact, there only one studies indicted the significant changes between shoes of HL and LL while other their studies noted insignificant difference was found for different type of shoes (Lam et al. 2015).

Figure 4 

Pooled results at ankle inversion changes between the high-top shoe and low-top shoe during the landing after jumping.

Figure 5 

Pooled results at ankle ROM changes between the high-top shoe and low-top shoe during touchdown after side-step cutting.

4. Discussion

This is the first review to evaluate the effect of shoe collar height on ankle stability in basketball players. The pooled results from the meta-analysis indicated that there no difference was found between the high-lop shoes and low-lop shoes in ankle inversion and ROM, especially in the motion of jumping and sidestep cutting task.

Due to the varies parameters and difference testing motion were included in 6 articles, the ankle inversion angle at landing after jumping and the ankle of ROM at sidestep cutting were respectively extracted in order to maintain the consistency of selected indicators to reduce the interference of other factors, while the parameters in these two conditions are more tested and can be conducted the meta-analysis. The current analysis is based on the comprehensive seek with accurate statistical calculation. The two studies have shown the significant difference in ankle inversion between the HL and LL (Brizuela et al. 1997; Ricard, Schulties, and Saret 2000), while the results of another research were inconsistent (Fu et al. 2014). However, through the strict statistical analysis indicated that HL was not superior to the LL in keeping the ankle stability. Rovere et al. partially testified this result as well when contrasted the effect of Hl shoes versus LL shoes for ankle sprain and failed to measure a significant difference in ankle inversion from the 46 basketball players (Rovere et al. 1988). In addition, a comparative analysis with respects to the effect of shoe type also showed no difference between high-top shoes and low-top shoes, especially those with no history of ankle injury (Gross, Liu, and therapy 2003). Then Robert et al. demonstrated whether maximal inversion and eversion moment can be altered by the collar height of a basketball shoe, the subjects wearing a low or a high-top basketball shoes were measured for a functional ankle strength testing with three different weight-bearing conditions 0°, 16°, and 32° respectively, in terms of shoe height the result showed that there a significant increase the resistance to the inversion moment were found in moderate ankle plantar flexion (0° and 16°) (Ottaviani et al. 1995). But in this research, it may be due to the limitation of experimental equipment led to authors only focused on the static neutral position, as we all know the main function of any type ankle support is strength ankle stability in basketball movement. And ankle inversion mainly happens in dynamic motion and is accompanied by ankle flexion (Wright et al. 2000), such as contacting on the surface after running, jumping, or cutting, instead of keeping in a static and neutral foot position condition (Robbins, Waked, and Rappel 1995). Thus, this conclusion does not represent the effect of shoe height on ankle in the frontal plane during the basketball motion. There are a lot of previous studies paying close attention to the stabilizing effect of HL shoes which was measured during the jumping and cutting motions (Stacoff et al. 1998). Brizuela et al. found that HL shoe led the greater peak inversion angles at landing, and it actually limited the ankle motion by increasing restriction in ROM, also weaken player performance in vertical jumping and in side-step cutting running (Brizuela et al. 1997). And it was suggested that HL shoe should be used to perform higher jump frequency to reduce the risk of injury because of the restrictive effect of collar height on ankle joint. In terms of the LL shoe, it was recommended for basketball players with few jumps, for those who need conduct more rapid movements frequently.

On the other hand, there is disagreement on the impact high-top shoes on limiting ankle inversion ROM, and few biomechanical studies have been conducted to solve this problem (Ashton-Miller et al. 1996; Verhagen, van der Beek, and van Mechelen 2001). Actually, according to researchers noted that this limitation effect is smaller, and this kind of restrictive impact of basketball shoes cannot incorporate sports activities and lowered the sport performance. Through the meta-analysis noted that there no difference in ankle inversion ROM between the HL and LL were found. This result was consistent with collection research by Fu et al. (Fu et al. 2014). Due to the different test methods were used in the collection papers, the results of those research indicated the different conditions. Some studies presented that the muscle activation strategy should be studied by electromyography (EMG) in the further to understand the underlying muscle activity pattern when wearing the different shoe collar in basketball players. Unfortunately, there few studies to observe the muscle activity of lower extremity between the HT shoes and LT shoes from the EMG perspective, thus there is not enough research to support the mete-analysis. But Fu et al. indicated the muscle activation of lower leg which included the tibialis anterior, peroneus longus, and peroneus brevis, the results showed that a pre-activation timing has been delayed and the magnitude of evertor muscle activity has been decreased in high-top shoe condition. It was suggesting that those changes in muscles may have a positive effect on ankle joint stability.

There are some limitations to this systematic review. Researches of selection studies with different testing methods and diverse experiment device, in which each extracting paper included different parameters, the two parameters of ankle inversion and ROM was presented at most of articles. Thus, it was limited to extract other data to analyze the effect of different types of shoes in ankle stability. In addition, the effects of shoe collar on muscle in the lower extremity were not concluded due to the only one study referred to this. Finally, the data was extracted from the final articles for comparison was not originally available, may lead to other bias.

5. Conclusion

In this review with meta-analysis of researches of ankle stability which includes ankle inversion and ROM in high-top shoes and low-top shoes during the movement of landing after jumping and touchdown after side-step cutting. The results showed that there no significant difference in shoe collar was found. Thus, the more parameters of ankle changes and muscle activity of lower extremity should be assessed in the future to improve our comprehensive understanding for shoe collar effect of ankle stability.

Competing Interests

The author has no competing interests to declare.


  1. Ashton-Miller, J. A., Ottaviani, R. A., Hutchinson, C., & Wojtys, E. M. (1996). What best protects the inverted weightbearing ankle against further inversion? Evertor muscle strength compares favorably with shoe height, athletic tape, and three orthoses. The American journal of sports medicine, 24(6), 800–809. DOI: 

  2. Barrett, J. R., Tanji, J. L., Drake, C., Fuller, D., Kawasaki, R. I., & Fenton, R. M. (1993). High-versus low-top shoes for the prevention of ankle sprains in basketball players: a prospective randomized study. The American Journal of Sports Medicine, 21(4), 582–585. DOI: 

  3. Blache, Y., Beguin, A., & Monteil, K. (2011). Effects of various parameters of basketball shoes on vertical jumping performance: A case study. Science & Sports, 26(1), 48–50. DOI: 

  4. Borenstein, M., Hedges, L., & Rothstein, H. (2007). Meta-analysis: Fixed effect vs. random effects. 

  5. Brizuela, G., Llana, S., Ferrandis, R., & Garcia-Belenguer, A. C. (1997). The influence of basketball shoes with increased ankle support on shock attenuation and performance in running and jumping. Journal of Sports Sciences, 15(5), 505–515. DOI: 

  6. Fu, W., Fang, Y., Liu, Y., & Hou, J. (2014). The effect of high-top and low-top shoes on ankle inversion kinematics and muscle activation in landing on a tilted surface. Journal of foot and ankle research, 7(1), 14. DOI: 

  7. Greene, A. J., Stuelcken, M. C., Smith, R. M., & Vanwanseele, B. (2014). The effect of external ankle support on the kinematics and kinetics of the lower limb during a side step cutting task in netballers. BMC Sports Science, Medicine and Rehabilitation, 6(1), 42. DOI: 

  8. Gross, M. T., & Liu, H. Y. (2003). The role of ankle bracing for prevention of ankle sprain injuries. Journal of orthopaedic & sports physical therapy, 33(10), 572–577. DOI: 

  9. Handoll, H. H., Rowe, B. H., Quinn, K. M., & de Bie, R. (2001). Interventions for preventing ankle ligament injuries. Cochrane Database of Systematic Reviews, (3). DOI: 

  10. Higgins, J. P., Thompson, S. G., Deeks, J. J., & Altman, D. G. (2003). Measuring inconsistency in meta-analyses. Bmj, 327(7414), 557–560. DOI: 

  11. Kerr, R., Arnold, G. P., Drew, T. S., Cochrane, L. A., & Abboud, R. J. (2009). Shoes influence lower limb muscle activity and may predispose the wearer to lateral ankle ligament injury. Journal of orthopaedic research, 27(3), 318–324. DOI: 

  12. Knight, A. C., & Weimar, W. H. (2011). Effects of inversion perturbation after step-down on the latency of the peroneus longus and peroneus brevis. Journal of applied biomechanics, 27(4), 283–290. DOI: 

  13. Lam, G. W. K., Park, E. J., Lee, K. K., & Cheung, J. T. M. (2015). Shoe collar height effect on athletic performance, ankle joint kinematics and kinetics during unanticipated maximum-effort side-cutting performance. Journal of sports sciences, 33(16), 1738–1749. DOI: 

  14. Liberati, A., Altman, D. G., Tetzlaff, J., Mulrow, C., Gøtzsche, P. C., Ioannidis, J. P., & Moher, D. (2009). The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. PLoS medicine, 6(7), e1000100. DOI: 

  15. Liu, H., Wu, Z., & Lam, W. K. (2017). Collar height and heel counter-stiffness for ankle stability and athletic performance in basketball. Research in Sports Medicine, 25(2), 209–218. DOI: 

  16. Mrdakovic, V., Ilic, D. B., Jankovic, N., Rajkovic, Z., & Stefanovic, D. (2008). Pre-activity modulation of lower extremity muscles within different types and heights of deep jump. Journal of sports science & medicine, 7(2), 269. 

  17. Neptune, R. R., Wright, I. C., & Van Den Bogert, A. J. (1999). Muscle coordination and function during cutting movements. Medicine and science in sports and exercise, 31, 294–302. DOI: 

  18. Ottaviani, R. A., Ashton-Miller, J. A., Kothari, S. U., & Wojtys, E. M. (1995). Basketball shoe height and the maximal muscular resistance to applied ankle inversion and eversion moments. The American Journal of Sports Medicine, 23(4), 418–423. DOI: 

  19. Ramanathan, A. K., Wallace, D. T., Arnold, G. P., Drew, T. S., Wang, W., & Abboud, R. J. (2011). The effect of varying footwear configurations on the peroneus longus muscle function following inversion. The Foot, 21(1), 31–36. DOI: 

  20. Ricard, M. D., Schulties, S. S., & Saret, J. J. (2000). Effects of high-top and low-top shoes on ankle inversion. Journal of athletic training, 35(1), 38. 

  21. Robbins, S., Waked, E., & Rappel, R. (1995). Ankle taping improves proprioception before and after exercise in young men. British Journal of Sports Medicine, 29(4), 242–247. DOI: 

  22. Rovere, G. D., Clarke, T. J., Yates, C. S., & Burley, K. (1988). Retrospective comparison of taping and ankle stabilizers in preventing ankle injuries. The American journal of sports medicine, 16(3), 228–233. DOI: 

  23. Ryan, L. (1994). Mechanical stability, muscle strength and proprioception in the functionally unstable ankle. Australian Journal of Physiotherapy, 40(1), 41–47. DOI: 

  24. Smith, R. W., & Reischl, S. F. (1986). Treatment of ankle sprains in young athletes. The American journal of sports medicine, 14(6), 465–471. DOI: 

  25. Stacoff, A., Avramakis, E., Siegenthaler, R., & Stüssi, E. (1998). High-cut shoes and lateral heel stability during cutting movements in floorball. Journal of Biomechanics, 1001(31), 178. DOI: 

  26. Verhagen, E. A., van der Beek, A. J., & van Mechelen, W. (2001). The effect of tape, braces and shoes on ankle range of motion. Sports Medicine, 31(9), 667–677. DOI: 

  27. Wright, I. C., Neptune, R. R., van den Bogert, A. J., & Nigg, B. M. (2000). The influence of foot positioning on ankle sprains. Journal of biomechanics, 33(5), 513–519. DOI: