**About the Editors**

#### **Jiwu Chen**

Professor Jiwu Chen, Ph.D, currently works in Shanghai General Hospital, Shanghai Jiaotong University, China. He has focused on clinical and basic research into sports injuries for a long time, and has been funded by the National Natural Science Foundation of China six times. He has more than 100 publications, among which several have been published in *The American Journal of Sports Medicine*, *Bioactive Materials*, *Small*, etc. Currently, Prof. Chen serves as the chairman-elect of the Sports Medicine Specialty Branch of the Shanghai Medical Association, the vice chairman of the Sports Medicine Branch of the Chinese Research Hospital Association, and the deputy leader of the Sports Medicine Group of the Orthopedic Branch of the Chinese Medical Doctor Association, as well as being a member of the ISAKOS Shoulder Joint Committee, etc. Prof. Chen's achivements in arthroscopic surgery are substantial, with 500–1000 surgeries per year. At the 2017 ISAKOS Biennial Conference, he served as the only invited Chinese expert to give a demonstration of arthroscopic surgery of a massive rotator cuff tear.

#### **Yaying Sun**

Dr. Yaying Sun, Ph.D, currently works in Huashan Hospital, Fudan University, Shanghai, China. He serves as the member of various editorial boards, including the *International Journal of Nanomedicine*, *Burns & Trauma*, *Journal of Clinical Medicine*, etc., and the reviewer of several scientific journals. In 2021, Dr. Sun received funding from the National Natural Science Foundation of China. His ongoing research interests are regenerative medicine of skeletal muscle and tendons/ligaments and the mechanism of aging of the musculoskeletal system, as well as multi-discipline research regarding the crosstalk between the musculoskeletal system, sports activity and exercise, and other disorders, especially cancer and chronic diseases.

### *Editorial* **Advancements in Sports Medicine**

**Yaying Sun <sup>1</sup> and Jiwu Chen 2,\***


Sports medicine has developed rapidly in recent years. Countless advancements have been achieved regarding the mechanism, repair, and recovery of sports injuries. The interaction between sports medicine and other disciplines is also a trending topic. Multi-disciplinary outcomes are being increasingly yielded. In this Special Issue, we have collected several clinical advancements in sports medicine.

Rotator cuff tear is a common shoulder disorder in clinical practice. Based on tear size, treatment methods can be divided into direct repair using suture anchors or repair with augmentation due to a massive tear size. Li et al. conducted a review of the current literature and reported that bone quality, insertion depth, insertion angle, size of rotator cuff tear, preoperative corticosteroid injections, anchor design, and the materials used to produce anchors may influence the anchor pullout strength, leading to a poor recovery [1]. Regarding massive rotator cuff tear that cannot be fixed directly, Wellington et al. used a biologically enhanced demineralized bone matrix for the augmentation and found that 10 of 20 patients who received this treatment still suffered from a retear at follow-up [2]. This outcome suggested that there is still a long way to go to enhance the repair of rotator cuff.

For trunk, spinal fusion is usually applied for patients with lumbar degeneration with overall good results. However, some may still experience failure. Guo et al. noticed that the albumin-to-alkaline phosphatase ratio can be used as a prognostic biomarker for measuring clinical outcomes after spinal fusion [3]. Regarding lower limb, Cong et al. reported a modified capsulotomy approach to facilitate the arthroscopic femoroplasty and acetabular labrum repair, with the clinical data supporting its popularization [4]. Regarding knee, Zhang et al. found that anterior cruciate ligament reconstruction using an insertion preservation technique has a protective effect on cartilage degeneration in long-term follow-up [5].

In addition to the general population, novel advancements have also been achieved regarding injuries in athletes. Martins established a predictive model for injury risk in football players [6], while Keller et al. pointed out some divergences in terms of exaggerated blood pressure response in athletes [7]. Gaudette et al. studied runner injuries and emphasized the importance of gait retaining for post-injury recovery [8]. Merle et al. focused on the oral health of young athletes and revealed an association of blood/performance indexes and periodontal inflammation [9].

Some interesting multi-disciplinary research is also included in this Issue. Guo et al. conducted a systematic review and found that aerobic plus machine-assisted resistance training may improve the vascular function in patients with type 2 diabetes [10], while Ma et al. found that adaptive posture-balance cardiac rehabilitation exercise could remarkably restore physical tolerance in a population suffering from cardiovascular diseases [11]. The interaction between sports medicine and other subjects will no doubt make this discipline more meaningful in the future.

**Author Contributions:** Y.S. organized the paper and J.C. revised the paper. All authors have read and agreed to the published version of the manuscript.

**Citation:** Sun, Y.; Chen, J. Advancements in Sports Medicine. *J. Clin. Med.* **2023**, *12*, 3489. https:// doi.org/10.3390/jcm12103489

Received: 3 April 2023 Accepted: 22 April 2023 Published: 16 May 2023

**Copyright:** © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


**Disclaimer/Publisher's Note:** The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

## *Review* **Risk Factors and Corresponding Management for Suture Anchor Pullout during Arthroscopic Rotator Cuff Repair**

**Xiangwei Li †, Yujia Xiao †, Han Shu, Xianding Sun and Mao Nie \***

Center for Joint Surgery, Department of Orthopaedics, The Second Affiliated Hospital of Chongqing Medical University, Chongqing 400010, China

**\*** Correspondence: 302218@cqmu.edu.cn

† These authors contributed equally to this work.

**Abstract:** Introduction: Due to the aging of the population, the incidence of rotator cuff tears is growing. For rotator cuff repair, arthroscopic suture-anchor repair has gradually replaced open transosseous repair, so suture anchors are now considered increasingly important in rotator cuff tear reconstruction. There are some but limited studies of suture anchor pullout after arthroscopic rotator cuff repair. However, there is no body of knowledge in this area, which makes it difficult for clinicians to predict the risk of anchor pullout comprehensively and manage it accordingly. Methods: The literature search included rotator cuff repair as well as anchor pullout strength. A review of the literature was performed including all articles published in PubMed until September 2021. Articles of all in vitro biomechanical and clinical trial levels in English were included. After assessing all abstracts (*n* = 275), the full text and the bibliographies of the relevant articles were analyzed for the questions posed (*n* = 80). Articles including outcomes without the area of interest were excluded (*n* = 22). The final literature research revealed 58 relevant articles. Narrative synthesis was undertaken to bring together the findings from studies included in this review. Result: Based on the presented studies, the overall incidence of anchor pullout is not low, and the incidence of intraoperative anchor pullout is slightly higher than in the early postoperative period. The risk factors for anchor pullout are mainly related to bone quality, insertion depth, insertion angle, size of rotator cuff tear, preoperative corticosteroid injections, anchor design, the materials used to produce anchors, etc. In response to the above issues, we have introduced and evaluated management techniques. They include changing the implant site of anchors, cement augmentation for suture anchors, increasing the number of suture limbs, using all-suture anchors, using an arthroscopic transosseous knotless anchor, the Buddy anchor technique, Steinmann pin anchoring, and transosseous suture repair technology. Discussion: However, not many of the management techniques have been widely used in clinical practice. Most of them come from in vitro biomechanical studies, so in vivo randomized controlled trials with larger sample sizes are needed to see if they can help patients in the long run.

**Keywords:** rotator cuff tear; rotator cuff repair; bone quality; osteopenia; osteoporosis; anchor pullout; pullout strength

#### **1. Introduction**

Due to the aging of the population, the incidence of rotator cuff tears is growing [1,2]. For rotator cuff repair, arthroscopic suture-anchor repair has gradually replaced open transosseous repair, so suture anchors are now considered increasingly important in rotator cuff tear reconstruction [3]. The majority of patients with rotator cuff tears are over 60 years old, and osteoporosis is very common among them [4,5]. This means that their proximal humeral bone quality is often poor, which will increase the incidence of anchor pullout [6,7] (Figure 1). Anchor pullout is one of the mechanisms of suture anchor failure. It occurs at the anchor-bone interface during arthroscopic rotator cuff repair, resulting in pullout of the anchors from the bone [8]. In terms of biomechanics, pullout strength is the pullout

**Citation:** Li, X.; Xiao, Y.; Shu, H.; Sun, X.; Nie, M. Risk Factors and Corresponding Management for Suture Anchor Pullout during Arthroscopic Rotator Cuff Repair. *J. Clin. Med.* **2022**, *11*, 6870. https://doi.org/10.3390/ jcm11226870

Academic Editor: Alessandro Castagna

Received: 6 September 2022 Accepted: 16 November 2022 Published: 21 November 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

force measured when anchor pullout occurs at the anchor-bone interface. Studies have attempted to find new methods to improve pullout strength, thus reducing the risk of anchor pullout.

**Figure 1.** (**a**) Intraoperative metallic suture anchor pullout; (**b**) intraoperative polyetheretherketone suture anchor pullout.

The purpose of this review is to summarize the body of knowledge on suture anchor pullout during arthroscopic rotator cuff repair. We will first briefly introduce the incidence of anchor pullout before discussing the reason why the suture anchor pulls out. Lastly, we will describe in detail the different technologies and studies that have been used to solve the problem of anchor pullout, and we will compare their pros and cons to help future practice.

#### **2. Method**

The literature search included rotator cuff repair as well as anchor pullout strength. A review of the literature was performed, including all articles published in PubMed until September 2021. The following search terms were used alone and in combination: "Rotator cuff repair", "anchor pullout", and "pullout strength". The articles were assessed considering the following research aspects: definition, incidence of anchor pullout, risk factor for anchor pullout, and anchor pullout management.

Articles of all in vitro biomechanical and clinical trial levels in English were included. After assessing all abstracts (*n* = 275), the full text and the bibliographies of the relevant articles were analyzed for the questions posed (*n* = 80). Articles including outcomes without the area of interest were excluded (*n* = 22). Specifically, publications from 1990 to 2021 were included because of the advancements in biomechanics, surgical treatments, and improved understanding of pullout strength for suture anchors. The final literature research revealed 58 relevant articles. Narrative synthesis was undertaken to bring together the findings from studies included in this review.

#### **3. Results**

#### *3.1. Total Incidence of Anchor Pullout*

The incidence of anchor pullout varies depending on the circumstances. The incidence of early anchor pullout after arthroscopic rotator cuff repair is approximately 0.1%–3.1%, while the incidence of anchor pullout during surgery is higher, approximately 3.3%–5.4%. Anchor pullout is one of the three mechanical mechanisms of revision surgery failure, with an incidence of approximately 4.5%.

#### 3.1.1. Early Anchor Pullout

A retrospective, monocentric study [9] by Skaliczki et al., showed that early anchor pullout was observed in six patients out of 5327 (0.1%).

Earlier studies found much higher rates of early anchor pullout. Benson et al. [10] investigated 269 patients who underwent arthroscopic rotator cuff repair and found six cases of early anchor pullout (2.4%). In their study of 127 patients, Dezaly et al. [11] reported a 3.1% prevalence of early anchor displacement. The difference may be caused by the different time points of the radiographic evaluation: The time period considered in the latter two studies also included part of the rehabilitation, while the observation time of Skaliczki et al. [9] was immediately after surgery. That is, early anchor pullout in the series of Skaliczki et al. [9] is mainly attributed to surgical intervention, while the results of Benson et al. [10] and Dezaly et al. [11] are at least partly due to the rehabilitation process.

#### 3.1.2. Anchor Pullout during Surgery

The incidence of anchor pullout during surgery has not been investigated, and only a few articles [9–11] have reported the incidence of early postoperative anchor failure.

A retrospective study [12] by Jung and colleagues showed that of 1076 patients who underwent arthroscopic rotator cuff repair, 483 were treated with screw-in-type bioabsorbable or biocomposite anchors, and 593 were treated with soft anchors. In the screw-in-type anchor group, 16 patients (3.3%, 16/483) experienced anchor pullout during surgery. Of the 593 patients treated by soft anchor insertion, 32 (5.4%, 32/593) experienced anchor pullout. These rates are not significantly different. Intraoperative anchor pullout was much more likely to happen in patients with larger rotator cuff tears, women, older people, or those who had shoulder stiffness before surgery.

#### 3.1.3. Anchor Pullout Has a Relatively Low Incidence

Cummins et al. [8] found three mechanisms by which rotator cuff repairs fail mechanically at the time of revision surgery: rotator cuff suture pullout from the repaired tendon (86.3% of cases), new tears in a different place (9.1% of cases), and complete anchor displacement, or pullout, from the bone (4.5% of cases).

#### *3.2. Risk Factors for Suture Anchor Pullout*

The reason why the suture anchor pulls out is the poor stability of anchor fixation in arthroscopic rotator cuff repair. The stability is determined by bone quality, insertion depth, insertion angle, anchor design, the materials used to produce anchors [13–19] and so on.

#### 3.2.1. Bone Quality

Pullout strength depends on bone mineral density [4,20,21].

Suture anchors have better pullout characteristics when placed in areas of higher bone mineral density (BMD) [4,20,22,23]. However, the use of anchors in patients who are elderly and who may be osteoporotic [24] can potentially increase the likelihood of anchor pullout. The bone quality of the greater tuberosity is one of the factors affecting repair integrity [6]. In patients with poor bone quality, the failure rate after rotator cuff repair is as high as 68% [4,25]. The quality of the proximal humerus bone also deteriorates with age and is more pronounced in patients who have RCT. Djurasovic et al. [7] reviewed 80 cases of failed rotator cuff repair and showed that 10% of them had anchor migration or loosening. Anchor migration is a state between anchor loosening and pullout, and it is incomplete anchor pullout. From these results, we can see that the lower the bone mineral density, the more easily the anchor will cut out of the humerus [4]. These studies provide a theoretical basis for various augmentation technologies.

#### 3.2.2. Anchor Material and Design

The mechanical fixation (pullout strength) of suture anchors is determined by their design, such as the pitch and number of threads, length, size, and overall shape [4,10,26]. Anchors of various designs, materials, and sizes have been invented. The pullout strength can differ according to the material and design of suture anchors [27–29].

Anchors made of different materials have different incidences of anchor pullout. Tingart et al. [4] found that the pullout strength of metal screw-type anchors is higher than that of biodegradable hook-type anchors. In addition, most studies using radiotransparent (RT) anchors have reported complications caused by bioabsorbable anchors resulting in bone lysis, defects, and sometimes fractures [30–32], which may lead to late pullout. However, osteolysis has no effect on clinical outcomes [33]. Polyetheretherketone (PEEK) anchors, although non-absorbable, also enlarge the peripheral bones significantly more laterally than medially in double-row (DR) repairs [34], which may be the reason why they pull out. In short, the use of anchors of different materials will produce different pullout strengths, which is the reason why anchor pullout occurs under specific circumstances.

A study by Chae and colleagues [35] indicated that high pullout strength was primarily attributed to geometric design factors of the suture anchors, such as greater contact surface area between the anchor threads and surrounding bone, overall length, number of threads, and height of the thread. It is possible that the contact surface area between the anchor threads and surrounding bone is related to other geometric design factors of the suture anchor, such as the overall length, diameter, number of threads, height of the thread, pitch, and helix angle. Chae et al., found that the number and height of threads were positively correlated with the pullout strength among suture anchors of several geometric designs. In fact, the number and height of threads are the most important geometric design factors for increasing the contact surface area between suture anchors and surrounding bone. Their results support the points of view that greater thread-to-bone surface contact leads to greater pullout strength and that screw threads impart improved holding strength due to the increased through contact with the surface of the bone [36–38]. Kang et al. [39] reported that a micropore bioabsorbable suture anchor had higher pullout strength, which may have been related to the bone growth induced by the micropore bioabsorbable anchor. Even though more clinical trials need to be conducted to confirm the above assumptions, there is no doubt that these research results point us in a new direction when it comes to anchor design.

Therefore, the clinical application of anchors with different designs will often bring about different pullout strengths and lead to different incidences of anchor pullout. Anchor design determines the stability of anchor fixation, which is one of the reasons why anchor design affects the incidence of anchor pullout.

#### 3.2.3. Number of Anchors (Distance between Anchors)

The relationships of the pullout strength to the anchor material, anchor design, insertion angle, insertion depth, and bone mineral density have been investigated [3,15,16,20,40–42]. However, these studies only focused on the pullout strength of one anchor. One study investigated the pullout strength of two anchors instead of one [43], finding that the pullout strength of two anchors was higher. Kawakami and colleagues [44] showed that in polyurethane and porcine models, the minimum distance between anchors to not reduce the pullout strength was 6 mm, which was less than the previously determined 10-mm separation, and this result was not affected by the different bone qualities, even when applied to osteoporotic bone. When two anchors are placed 4 mm apart, there are two possible reasons for the decrease in pullout strength. First, when two anchors are very close, the cancellous bone around the anchors is not strong enough to support both. Second, in fact, a 4-mm distance means that adjacent anchors will overlap. The contact area between the anchor thread and the cancellous bone decreases as the amount of overlap increases. The contact area of the anchor thread is closely linked to the pullout strength [37,45]. When calculating the distance between two anchors, we mean from center to center, so the minimum distance without decreasing the pullout strength may be different for suture anchors with different diameters. However, due to financial constraints, only two types of suture anchors were examined in this study, compared to the ideal situation of testing all commercially available anchors.

All-suture anchors are biomechanically inferior to screw-in-type anchors [46]. However, Ntalos et al. [47] reported that all-suture anchors and traditional anchors had similar

average pullout strengths in an unlimited cyclic model. Moreover, compared with traditional anchors, all-suture anchors have a smaller volume, which allows more of them to be implanted in the same volume of bone [48]. The overall biomechanical performance is improved by sharing the load at multiple fixation points. However, the minimum distance between all-suture anchors seems not to have been reported.

#### 3.2.4. Insertion Angle

Accidental anchor pullout is a common mechanism of repair failure, and its occurrence is affected by bone quality and the implantation technique [49,50]. The relationship between anchor pullout and anchor insertion angle has also been studied. It was widely accepted and understood that placing the anchor at 45◦ to the insertion surface would display the strongest pullout strength [4,15,18–20,51] after Burkhart's proposal of the deadman theory in 1995 [52].

In 2009, Strauss et al., used cadaveric shoulders to study the effects of anchor insertion angle and rotator cuff tendon repair [49]. The torn supraspinatus tendons were repaired by single suture anchor with an insertion angle of 45◦ or 90◦. The results showed that the rotator cuff repair with the anchor inserted at 90◦ to the bone surface was stronger than the repair with the anchor inserted at 45◦. However, compared with the whole repaired construct, the effect of insertion angle on just anchor pullout strength was of more interest to the researchers.

In 2014, Clevenger et al., tested the pullout strength of anchors with insertion angles from 45◦ to 135◦ in 15◦ increments [14]. According to the findings, anchors set at an acute angle to the pulling axis were substantially weaker than those positioned at an obtuse angle. It did not appear to necessarily match the clinical settings, though, as just one type of synthetic cancellous bone was used and no cortical bone. In 2016, Nagamoto et al., conducted a biomechanical test of anchor insertion angle using the greater tuberosity of porcine humeri and three different densities of synthetic cancellous bones with a 2-mmthick cortical bone connected to one side. Their findings showed that regardless of bone density, the pullout strength of the anchors implanted at 90◦ to the bone surface was higher than the anchors inserted at 45◦ [16].

In the same year, Itoi et al., comprehensively evaluated their laboratory data against previous data and concluded that insertion angles of 45◦ and 90◦ were the strongest for threadless and threaded anchors, respectively [53,54]. So, whether threaded or threadless anchors are used should also affect the choice of insertion angle. Threadless anchors provide less friction. In this case, inserting an anchor at 45◦ had a higher pullout strength than inserting an anchor at 90◦ or more. In contrast, threaded anchors can provide substantial friction. Therefore, the maximum pullout strength can be obtained by inserting the anchor at 90◦.

In 2018, Ntalos et al.'s [47] biomechanical study demonstrated that the maximal force in all-suture and traditional anchors could be detected at a 90◦ insertion angle. Regardless of the kind, the pullout strength was decreased when they were inserted at more acute (45◦) or obtuse (110◦) angles. Those differences were not statistically significant, though. They thought that the angle at which the anchor was inserted was not as important in the clinic as people had thought [47].

#### 3.2.5. Size of Rotator Cuff Tear

The retrospective cohort study by Benson and colleagues provided conclusive evidence that patients with larger rotator cuff tears have a significantly higher incidence of anchor pullout. They found [10] that among 251 patients who used metallic suture anchor for rotator cuff repair, six had early anchor pullout, with an overall incidence of about 2.4%. The incidence of rotator cuff tears less than or equal to 3 cm was 0.5%, and the incidence in tears greater than 3 cm was 11%. In large tears, the suture anchor bears higher tension, so the incidence of anchor pullout will also be higher.

#### 3.2.6. Insertion Depth

In a biomechanical study, Bynum et al. [15] showed that changing the insertion depth of the suture anchor affected the mechanical properties and the failure modes of suture anchor constructs. Suture anchors inserted with the suture eyelet deep had premature failure because of construct elongation.

Kirchhof et al. [55] reported that screwing the anchor deeper did not increase the pullout strength. This is because the deep bone mineral density of the greater tuberosity is relatively low. For patients with osteoporosis, this is of no help. Osteoporosis usually involves the patient's cancellous bone first, resulting in a decrease in cancellous bone quality, followed by cortical bone. Therefore, the deep bone mineral density of the greater tuberosity for patients with osteoporosis is relatively low, and screwing the anchor deeper cannot improve the pullout strength.

Therefore, there is no consistent conclusion on whether increasing the insertion depth of the anchor improves the pullout strength.

#### 3.2.7. The Effect of Corticosteroid Injections on Anchor Pullout Strength

Because RCT patients usually have obvious pain symptoms, corticosteroid injections (CSIs) into the subacromial space have been an important treatment for RCT patients. Puzzitiello et al. [56] showed that for patients who had received CSIs within two weeks, their suture anchor pullout strength decreased significantly after arthroscopic rotator cuff repair. There was no significant decrease after 3 or 4 weeks. These findings suggest that for patients who have received CSIs before surgery, we should ensure that they receive surgery after a certain interval of time.

#### *3.3. Anchor Pullout Management*

As mentioned above, rotator cuff repair has a high retear rate, and the risk of failure increases with the age of the patient [57] and with the size of the tear [58]. The quality of the proximal humerus also deteriorates with age, and this phenomenon is more common in RCT patients. With the continuous development and popularity of arthroscopic rotator cuff repair, the practice of open rotator cuff repair in the new generation of surgeons is becoming rarer. This section focuses on various management techniques and biomechanical principles for anchor pullout during arthroscopic rotator cuff repair.

#### 3.3.1. Changing the Implant Site of Anchors

For arthroscopic rotator cuff repair, the suture anchor is implanted in the proximal humerus, usually into the greater tuberosity. Many studies have analyzed the bone quality distribution of the greater tuberosity.

In 2003, an in vitro biomechanical study by Tingart et al. [4] demonstrated that, within the proximal part of the greater tuberosity, trabecular bone mineral density of the posterior region and cortical bone mineral density of the middle region were highest, respectively. However, loads to failure in the anterior and middle regions were, on average, 62% higher than the load to failure in the posterior region. They came to the conclusion that cortical bone mineral density was a stronger predictor of pullout strength in the proximal region of the tuberosity than trabecular bone mineral density. The pullout strength might be improved by placing suture anchors in the proximal-anterior and proximal-middle regions of greater tuberosity [20].

Kirchhof et al. [5] performed high-resolution peripheral quantitative CT scanning on 36 cadaver specimens, finding that the volume of highest bone quality was found at the posteromedial aspect. Sakamoto et al. [59] used multidetector row computed tomography to successfully perform an in vivo evaluation of the bone microstructure of the humeral greater tuberosity in patients with rotator cuff tears. They also obtained the same results as Kirchhof et al. According to the findings of both studies, the posterior medial region of the greater tuberosity was the best location for anchor insertion in terms of bone quality. This contradicted the results of Tingart et al. [4].

#### 3.3.2. Cement Augmentation for Suture Anchors

Bone grafting or using bone cement to fill the void caused by osteoporotic bone resorption or large cystic changes within the subchondral plate can effectively improve the bone quality of patients undergoing arthroscopic rotator cuff repair. It is very difficult to perform structural bone grafting under arthroscopy, and the pullout strength will not be improved immediately, so it is clinically more feasible to inject bone cement to enhance bone quality and improve pullout strength.

Oshtory and colleagues [60] reported that the pullout strength of suture anchors injected with tricalcium phosphate cement increased by 29%. Giori and colleagues [61] reported a 71% gain in pullout strength with anchor augmentation by polymethyl methacrylate (PMMA) cement. Although the pullout strength was improved, PMMA cement is not bioabsorbable, which may make revision surgery harder. Moreover, PMMA cement produces a thermal effect during the curing process, which may also cause bone necrosis, making the pullout strength uncontrollable in specific cases.

Postl and colleagues [62] reported that the pullout strength of suture anchors injected with bioabsorbable and fiber-reinforced calcium phosphate cement increased by 66.8%. This fiber-reinforced calcium phosphate cement can reach a pullout strength similar to that of PMMA cement but also retains the properties of calcium phosphate cement; that is, it does not produce a thermal effect and is bioabsorbable [63]. This new bone cement combines the advantages of calcium phosphate cement and PMMA cement and is a promising reinforcing material. To be applied in the clinic, it needs to be evaluated in further in vivo experiments.

The biomechanical results above show that it is theoretically tenable to improve the pullout strength of different materials by bone cement augmentation (Table 1).

**Table 1.** Pullout strength increment for augmentation with different types of bone cement.


In fact, not only the material of the bone cement but also the injection method of bone cement has a great impact on the final biomechanical results. Braunstein et al. [64] drilled a hole first, then injected bone cement, and finally implanted an anchor. However, this method can easily lead to the extrusion of bone cement, which is not feasible in an arthroscopic setting. Aziz and colleagues [65] introduced a new bone cement injection method that used an open architecture-type anchor. This method allowed the operator to implant the anchor first and then directly inject bone cement through a cannulated in situ suture anchor with fenestrations. This anchor can make the bone cement interlace and bond with the surrounding bone better, increasing the surface area in contact with the bone, so it may have higher pullout strength. At the same time, we can limit the bone cement injection to the distal end of the anchor, which can effectively reduce the occurrence of bone cement extrusion and the thermal effect on the healing surface, thereby reducing the

incidence of bone necrosis. In this way, the method can help to retain the bone quality and improve the pullout strength.

The experimental results of in vitro biomechanical studies also confirm the hypothesis above. Aziz and colleagues [65] reported that the pullout strength increased by 167% when bone cement was injected through an open architecture-type anchor, which was much higher than the pullout strength obtained by using the injection method of Braunstein et al. [64], which was only 45% to 47%, depending on the anatomic location.

#### 3.3.3. Using All-Suture Anchors

The pullout strength of an all-suture anchor mainly depends on the thickness of cortical bone [66]. Therefore, preoperative cortical bone thickness evaluation and no decortication during operation are particularly important to improve the pullout strength of all-suture anchors.

There is controversy about comparisons of pullout strength between all-suture anchors and traditional anchors. Negra and colleagues found that the failure load of all-suture anchors is less than that of traditional anchors, and they also have a significantly greater rate of anchor pullout by various failure mechanisms than traditional anchors [46]. However, this conclusion still needs to be further verified in a representative repair model. On the contrary, Ntalos et al. [47] confirmed that all-suture anchors and conventional anchors have no significant difference in biomechanical effects, and their pullout strength is also similar.

All-suture anchors have a much smaller volume than the traditional screw-in anchors, which allows us to enhance the pullout strength of the repaired construct by implanting more all-suture anchors in the limited bone [67].

#### 3.3.4. Increasing the Number of Suture Limbs

Shi et al. [68] found that when controlling for the number of sutures, using more suture limbs will result in a higher ultimate failure load. Conversely, when controlling for the number of sutured limbs, they found no significant differences among SR anchored, DR anchored, TOE, and transosseous repairs. In fact, they found that the number of sutures, the number of suture limbs, and the number of mattress stitches were more important in determining the overall strength than the suture structure.

#### 3.3.5. Buddy Anchor Technique

As we know, in patients with osteoporosis, the inserted suture anchors are likely to be unstable. Thus, Brady and Burkhart [36] introduced the buddy anchor technique as a salvage technique: a second anchor is inserted adjacent to the loose anchor to create an interference fit and subsequent higher pullout strength [36,38]. As reported by Denard and Burkhart, the essential mechanism of the buddy anchor system is reinforcement of the pull-out strength by interference fit [69].

One biomechanical study by Horoz et al., supported this technique [38]. They found that in osteoporotic bone, two interlocking suture anchors were stronger than a single anchor. The pullout strength was increased by interlocking a second suture anchor with the first. However, another study contradicted this finding [44]. The opposing view was that placing the two anchors to overlap would reduce the anchor bone contact area and thus reduce the pullout strength [44]. However, the effectiveness of the buddy anchor technique was for the original loose anchor, and the study did not evaluate whether the use of the buddy anchor technique helped to enhance the fixation of the original loose anchor. These two studies were in vitro biomechanical, and more in vivo clinical studies are needed to demonstrate the effect of buddy screwing in the future.

Jung et al. [12] used the buddy screwing technique to augment repair in 16 patients who experienced intraoperative anchor pullout. Three patients had early postoperative failure after buddy screwing. They thought that placing another anchor in an enlarged area tended to result in instability. However, this study was not strictly a randomized controlled trial. The number of cases was also limited since intraoperative anchor pullout was uncommon.

#### 3.3.6. Steinmann Pin Anchoring

Jung et al., invented a new anchor pullout management technique, bar anchoring with a threaded Steinmann pin (BASP) [12]. Using a threaded Steinmann pin (S-pin) (2.3 mm) and sutures, BASP was used to anchor pullouts during surgery. A threaded S-pin was trimmed to a length of 25–30 mm, and the center two-fifths of the S-pin were wrapped with three strands of a No. 2 high-strength suture and tied. A grasper was used to move the S-pin to the pullout site after its short part had been inserted through the anchor insertion portal. A specially made impactor was then attached to the end of the S-pin while being held below the cortical bone of the GT. The suture strands were then withdrawn to cause the S-pin to flip into the cancellous bone of the GT. Firm tension was gradually applied to the strands while observing the S-pin through the GT hole to ensure fixation. The three strands were attached to the S-pin using the Revo knot, a non-sliding knot, and were then used to repair the ruptured tendon.

In this study, the success rate of pullout management was 100% (13/13) for the BASP technique. At 6 months postoperatively, the tendon healing rate in patients undergoing BASP was 92.3% (12/13).

We can say that the BASP technique achieves satisfactory results both in terms of preventing suture anchor re-pullout and improving the tendon healing rate.

#### 3.3.7. Using an Arthroscopic Transosseous Knotless Anchor

For arthroscopic rotator cuff repair, the most commonly used technology is TOE repair technology. It is not a real transosseous repair technology, and suture anchors are still needed, which means that this technology still brings a risk of anchor pullout, especially in elderly patients and patients with osteoporosis.

Therefore, some surgeons have developed an arthroscopic transosseous knotless (ATOK) anchor to realize true transosseous repair through arthroscopic technology. A noninferiority trial by Sandow and colleagues [70] showed that none of the 15 patients who received the ATOK anchor for rotator cuff repair had anchor displacement or anchor pullout.

Compared with the widely used TOE repair technology, ATOK anchor repair can potentially reduce the incidence of anchor pullout. The effect of this technology also needs to be validated by randomized controlled trials with larger sample sizes.

#### 3.3.8. Transosseous Suture Repair Technology

The advent of suture anchors helped popularize arthroscopic rotator cuff repair due to the ease and speed of operation and their facilitation of instrumentation [20]. However, arthroscopic rotator cuff repair with suture anchors is not reliable for patients with osteoporosis, so some scholars have readopted transosseous suture repair. Of course, the current transosseous suture repair is not the same as the earlier arthrotomy but is performed under arthroscopy.

Randelli et al. [71] conducted a randomized controlled trial to compare the effectiveness of arthroscopic transosseous repair to single-row suture anchor repair. The two procedures produced equal results in terms of functional and radiological outcomes. Moreover, transosseous repair was found to reduce pain more quickly in the first month after surgery.

A matched cohort study by Srikumaran et al. [72] showed that in terms of patient-reported results, shoulder range of motion, and structural integrity, there are no differences between transosseous and transosseous equivalent suture-bridge rotator cuff repair procedures. The operating time was the same for all procedures. However, future randomized controlled trials are still needed to further demonstrate the equivalence of the two techniques.

These results demonstrate that arthroscopic transosseous repair can achieve the same results as arthroscopic rotator cuff repair with suture anchor in all aspects, and as a repair technique using only sutures without anchors, it can be used as an alternative treatment option for patients with anchor pullout. However, in patients with osteoporosis, suture cutting of the bone may also lead to failure of the repair.

In order to avoid bone cutting, some authors advise using a broader suture, such as a 2 mm tape, rather than the thinner No2 wire [73]. Due to its ideal viscoelastic properties [74] and broader contact surface with bone and soft tissue, this tape would exert the same force but less pressure at the contact region.

A cohort study by Beauchamp et al. [75] showed that arthroscopic transosseous repair using 2 mm tape material achieved significant mid-term functional improvement in this group of patients, with results statistically unaffected by larger tear size (>3 cm) or older age (≥65 years), which also happen to be risk factors for anchor pullout after rotator cuff repair with suture anchor. Therefore, arthroscopic transosseous repair using 2 mm braided suture tape could be an alternative surgical option to reduce the risk of anchor pullout for these two types of patients.

#### **4. Discussion and Clinical Inspirations**

This review organizes the body of knowledge on anchor pullout through a literature review. However, there are still some issues that need to be addressed in this area.

There are very few studies on the incidence of anchor pullout. Although the overall incidence of anchor pullout is not too high [8–12], studies of pullout rates under certain conditions, such as in patients with poor bone quality, needed, which are important for clinicians' preoperative decision making.

Additionally, we present the risk factors for anchor pullout. The effect of bone quality on pullout strength is relatively well established, and clinical studies with larger sample sizes would provide stronger support for the existing view. For anchor material and design, anchor insertion angle and depth, the existing findings are mainly from in vitro biomechanical studies [30–40,46–48], which do not fully simulate the clinical situation, and more in vivo clinical studies are needed to confirm the existing findings in the future. The minimum distance (center-to-center) of suture anchors without decreasing the pullout strength varies with anchor diameter [44]. The available studies only tested two types of suture anchors rather than all commercially available suture anchors. The minimum distance between all suture anchors has not been reported yet. Since the level of evidence from existing studies is low, future randomized controlled trials are needed to evaluate the effect of rotator cuff tear size and CSIs on pullout strength [10,56].

The focus of this review is corresponding management for suture anchor pullout. For the anchor implantation site, Kirchhof and Sakamoto only evaluated the distribution of total bone mineral density in the greater tuberosity without performing the corresponding biomechanical tests, and their findings cannot be used as predictors of anchor pullout strength [5,59]. From this point of view, the conclusions of Tingart et al. [4] seem to be more credible. Their study showed that placing the anchor in the anterior and middle regions proximal to the GT resulted in an average load to failure 62% higher than placing it in the posterior region [20]. Prospective clinical trials are necessary to understand whether the available managements can reduce anchor pullout rates and improve patient prognosis. By comparing the in vitro biomechanical data, we found that PMMA bone cement provides greater pullout strength compared to various new bone cements, despite its various drawbacks [60,62,63]. For the bone cement injection method, the injection of bone cement through an open architecture-type anchor is also superior to the traditional method of drilling a hole first, then injecting bone cement, and finally implanting the anchor [64,65]. However, the arthroscopic application of this technique still needs to overcome some technical difficulties, which require additional in vivo clinical studies. When a patient is at high risk for anchor pullout, we can assess their cortical bone thickness preoperatively, and if the cortical bone quality is good, we can use all suture anchors for rotator cuff repair because they are small and can be implanted in greater numbers, which can improve the overall pullout strength [66,67]. Buddy screwing, BASP, and ATOK are three relatively new techniques. They are not only theoretically valid but have also been demonstrated in

several studies. In fact, buddy anchor technique is a controversial technique. In the case of a small sample size, its in vivo application has a failure rate of 19% (3/16) [12], but it remains one of the few means of remedy in the event of intraoperative anchor pullout. In contrast, better pullout strength was achieved using the BASP technique and the ATOK anchor, and neither in vivo study reported anchor displacement or pullout [12,70]. However, these two techniques are more complex to perform, and in practice, buddy screwing remains a trusted and relatively simple remedy. However, neither of these studies were strict randomized controlled trials, nor was the sample size large enough [1,36,38,44,69,70]. Relatively speaking, arthroscopic transosseous suture repair is a more established technique. Its equivalence to suture anchor repair was also confirmed by a randomized controlled trial [71]. Arthroscopic transosseous suture repair does not involve the use of suture anchors at all, which is very suitable for patients at high risk of anchor pullout. However, this technique also presents a new problem; that is, the sutures may cut the osteoporotic bone, leading to repair failure. However, there are no studies on the probability of anchor pullout and bone cutting with suture anchor repair and transosseous suture repair for the same bone quality, respectively. Some cohort studies suggest that the use of wider sutures may reduce the risk of transosseous suture repair failure [73–75]. However, due to the uncertainty about the incidence of bone cutting, in vitro biomechanical studies using a severe osteoporosis model may help to increase positive results and thus help us better evaluate the effectiveness of this approach.

In conclusion, not many of the management techniques have been widely used in clinical practice. Since most are derived from in vitro biomechanical studies, in vivo randomized controlled trials with larger sample sizes are needed to confirm whether they can ultimately benefit patients.

**Author Contributions:** Conceptualization, X.L. and M.N.; methodology, X.L.; writing—original draft preparation, X.L. and Y.X.; writing—review and editing, X.L., Y.X., H.S., X.S. and M.N. All authors have read and agreed to the published version of the manuscript.

**Funding:** This work is supported by research funding from the National Natural Science Foundation of China (No.82002307), Natural Science Foundation of Chongqing (cstc2021jcyj-msxmX0121). Mao Nie and Xianding Sun was supported by the Kuanren Talents Program of the second affiliated hospital of Chongqing Medical University.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** No new data were created or analyzed in this study. Data sharing is not applicable to this article.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


### *Review* **Clinical Application of Gait Retraining in the Injured Runner**

**Logan W. Gaudette 1, Molly M. Bradach 1, José Roberto de Souza Junior 1,2, Bryan Heiderscheit 3, Caleb D. Johnson 1,4, Joshua Posilkin 1, Mitchell J. Rauh 5, Lauren K. Sara 1, Lindsay Wasserman 1, Karsten Hollander <sup>6</sup> and Adam S. Tenforde 1,\***


**Abstract:** Despite its positive influence on physical and mental wellbeing, running is associated with a high incidence of musculoskeletal injury. Potential modifiable risk factors for running-related injury have been identified, including running biomechanics. Gait retraining is used to address these biomechanical risk factors in injured runners. While recent systematic reviews of biomechanical risk factors for running-related injury and gait retraining have been conducted, there is a lack of information surrounding the translation of gait retraining for injured runners into clinical settings. Gait retraining studies in patients with patellofemoral pain syndrome have shown a decrease in pain and increase in functionality through increasing cadence, decreasing hip adduction, transitioning to a non-rearfoot strike pattern, increasing forward trunk lean, or a combination of some of these techniques. This literature suggests that gait retraining could be applied to the treatment of other injuries in runners, although there is limited evidence to support this specific to other running-related injuries. Components of successful gait retraining to treat injured runners with running-related injuries are presented.

**Keywords:** gait retraining; running-related injuries; kinetics; kinematics; rehabilitation

#### **1. Background**

The sport of running has positive effects on both physical [1] and mental [2] wellbeing. Unfortunately, runners experience a high rate of running-related injuries (RRIs). While reports of incidence rate vary depending on the population, up to 79% of recreational runners suffer a RRI each year [3]. In addition, RRIs have a high rate of recurrence. For example, female youth runners with a history of bone stress injury (BSI) have a 5 times elevated risk of sustaining a subsequent BSI [4]. Similar rates of injury reoccurrence were found in high school cross country runners [5].

Prior research has characterized risk factors for RRI. Of those that are modifiable, risk factors include neuromuscular, kinetic, kinematic, and spatiotemporal variables. While muscle weakness and imbalance may contribute to RRIs [6], strengthening alone may be insufficient for modifying biomechanical abnormalities that contribute to RRI [7]. Several variables related to running mechanics are thought to be related to injury and are frequent targets of intervention, including hip adduction [8,9], trunk lean [10,11], vertical loading rates [12–18], and step rate [19–22].

Gait retraining has been described as a method to change running biomechanics contributing to a given RRI [23]. Gait retraining using external feedback was first described

**Citation:** Gaudette, L.W.; Bradach, M.M.; de Souza Junior, J.R.; Heiderscheit, B.; Johnson, C.D.; Posilkin, J.; Rauh, M.J.; Sara, L.K.; Wasserman, L.; Hollander, K.; et al. Clinical Application of Gait Retraining in the Injured Runner. *J. Clin. Med.* **2022**, *11*, 6497. https:// doi.org/10.3390/jcm11216497

Academic Editors: Jiwu Chen and Yaying Sun

Received: 6 October 2022 Accepted: 31 October 2022 Published: 1 November 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

in patients following stroke [24]. It used concepts of motor learning for both acquisition and transfer phases and resulted in sustained improvements in hemiparetic gait. These concepts have since been applied to the management of RRIs by addressing a variety of aspects related to gait mechanics. Gait retraining typically involves the use of devices to measure the targeted biomechanical variable and provide external visual, verbal, or auditory cues to facilitate change. These external cues are described as biofeedback, and a faded biofeedback design refers to gradual reduction in external cues to promote learning of a new desired movement pattern without further feedback [24].

For gait retraining to be effective, biomechanical risk factors associated with RRI must be properly identified, and, if possible, addressed appropriately during treatment. The results of a recent systematic review and meta-analysis, which evaluated different forms of gait retraining [23], suggest that the literature surrounding gait retraining is occasionally inconsistent and largely inconclusive. There is need for a more easily digestible guide for clinicians seeking to implement gait retraining in the treatment of injured runners. The purpose of this narrative review is to provide a practical overview of what is known on biomechanical risk factors for RRI, gait retraining strategies to alter these risk factors, and provide clinical practical application of this knowledge.

#### **2. Methods**

This narrative review included studies related to the following topics: (I) Biomechanical risk factors for RRI, and (II) Gait retraining for runners with RRI. No restrictions were placed on language, publication date, participant age, gender, and duration of symptoms or stage of disease. Articles were excluded if: (I) running was not the primary focus of the study; (II) there was not a clear description of the gait retraining protocol used; (III) feedback was not removed after gait retraining to determine if gait adjustments could be maintained; (IV) the protocol did not use multiple sessions to allow for motor learning to occur. Additionally, studies that were not presented as a full manuscript (i.e., abstracts) were excluded. PubMed and EMBASE were the databases used. The date of the last search was 30 June 2022. Subject headings, synonyms, relevant terms, and variant spellings of three concepts (running biomechanics; gait retraining; running-related injuries) were used for the searches on each database. This strategy was used for each database with the appropriate truncation. All references were imported into Mendeley Reference Manager (Version 2.65.0), and duplicates were removed. The screening of eligible studies was performed in two steps: (I) screening the titles and abstracts, and (II) screening the full texts. List of references of the retrieved studies were searched to identify additional publications. Eligibility assessment was performed by two reviewers (LWG and MB). Disagreements were resolved by discussion between the two review authors. If no agreement could be reached, a final arbitration was performed by a third independent reviewer (AST). Relevant information was organized using the following topics: (I) Biomechanical risk factors for RRI; (II) Gait retraining overview; (III) Interventions characterizing gait retraining variables; (IV) Clinical application of gait retraining; (V) Limitation of current gait retraining strategies.

#### **3. Biomechanical Risk Factors for Running-Related Injury (RRI)**

Two recent systematic reviews of prospective studies have examined potential biomechanical risk factors for RRI [25,26]. Both reviews concluded that there was not strong evidence for a single biomechanical variable as a risk factor for all RRIs. The lack of an association between biomechanical variables and grouped RRI indicates the importance of investigating injury-specific biomechanical risk factors for RRI.

A recent systemic review examined biomechanical risk factors for several common RRIs including hamstring tendinopathy, patellofemoral pain syndrome (PFPS), patellar tendinopathy, iliotibial band syndrome, medial tibial stress syndrome, tibial stress fractures, Achilles tendinopathy, and plantar fasciitis [27]. Criteria for inclusion of a biomechanical risk factor in the study was a significant difference from a control group in one prospective

study or two retrospective studies [27]. Levels of evidence for biomechanical risk factors for specific RRIs varied from conflicting evidence to moderate evidence.

The strongest evidence supported decreased braking impulse [28,29] and increased ground contact time [28,29] for PFPS, increased duration of rearfoot eversion angle [30,31] and increased contralateral pelvic drop angle [30,32] for medial tibial stress syndrome, and increased average [33,34] and instantaneous loading rate [33,35] of vertical ground reaction force for plantar fasciitis. Each biomechanical variable was observed to have moderate evidence as risk factors for specific RRIs. More limited evidence was found for biomechanical risk factors for Achilles tendinopathy and tibial stress fractures, and very limited evidence was found for iliotibial band syndrome. No biomechanical variables met the study criteria for inclusion for patellar tendinopathy and hamstring tendinopathy. Definitions of potential biomechanical risk factors can be found below in Table 1.


**Table 1.** Definitions of key biomechanical variables.

#### **4. Gait Retraining Overview**

A narrative review published in 2020 evaluated the use of gait retraining as an intervention for PFPS [37]. The review included mostly case series or studies that did not contain a control group. Most biomechanical interventions included in the review, specifically decreasing hip adduction, increasing trunk lean, transitioning from a rearfoot strike (RFS) to a forefoot strike pattern, and increasing cadence, resulted in a reduction of pain [37]. The review also concluded that greater gait retraining session volume and a faded feedback design resulted in better outcomes compared to studies without a faded feedback design.

While a previous systematic review conducted in 2015 found foot strike manipulation had the greatest effect on kinematic measures and live feedback of tibial acceleration had the greatest effect on kinetic measures [38], a more recent systematic review and metaanalysis of gait retraining by Doyle et al. [23], which included only randomized controlled trials, concluded that the best evidence for gait retraining for runners supported step rate-based gait retraining. Though it achieved only moderate evidence, step rate-based gait retraining was shown to increase step rate, decrease stride length, decrease peak hip adduction (HADD) during stance, increase footstrike angle at initial contact and decrease VALR and VILR. Moderate evidence was also found for the ability of tibial acceleration based gait retraining to lower VILR.

While some gait retraining studies have reported a decrease in pain and improvement in functional outcomes in patients that underwent gait retraining, an insufficient number of studies reported pain measurements or clinical outcomes compared to a control group for inclusion in the systematic review and meta-analysis for conclusions regarding the effectiveness of gait retraining on patient pain or clinical outcomes to be made [23]. A summary of gait retraining studies and their effect on various clinical and biomechanical outcome variables can be seen in Table 2.


**Table 2.** Overview of Gait Retraining Studies.


#### **Table 2.** *Cont.*


**Table 2.** *Cont.*

Clinical Outcomes: LEFI- Lower Extremity Functional Index (same as LEFS), KOS-ADLS- Knee Outcome Survey– Activities of Daily Living Scale, SANE- Single Assessment Numeric Evaluation, LLOS- Lower Leg Outcome Survey, PSC- Patient Specific Complaints questionnaire, LEFS- Lower Extremity Functional Scale, AKPS- Anterior Knee Pain Scale, VAS- Visual Analog Scale.

#### **5. Interventions Characterizing Gait Retraining Variables**

Biomechanical targets of gait retraining studies (shown above in Table 2) include degree of hip adduction [8,9], step rate [19–21], tibial acceleration [12,14–18,39,45], footstrike pattern [41–44] and trunk lean [10]. The outcome measures of each study varied and included pain and functionality in injured runners, joint angles, and measures of loading rates (such as VALR and VILR) in healthy runners.

While the literature supporting hip adduction as a risk factor for PFPS is inconsistent [27], two studies have reported beneficial outcomes related to reduced hip adduction in runners with PFPS [8,9]. One study used a real time display generated by motion capture technology to display hip adduction [8], while the other study used visual feedback with a full-length mirror [9]. Both studies reported decreases in pain and increases in functionality, exceeding the minimal clinically important difference, at the end of the trial and at a 1- or 3-month follow-up.

Increases in step rate by 5% or 10% have been shown to lower COM vertical excursion, decrease breaking impulse and transition runners towards a more anterior footstrike pattern [36]. Three gait retraining studies have shown that step rate can be altered through gait retraining outside of a laboratory setting, including while runners continue with their training [20,21,50]. Willy et al. [20] found that this increase in step rate was also associated with a decrease in VALR, VILR and HADD [20]. A decrease in pain and increase in functionality was seen at the 6-month follow-up in runners that underwent gait retraining to increase their step rate by 10% [11]. The changes in step rate seen after gait retraining may not be constant at faster speeds, however [50]. One prospective study found no difference in KOS-ADLS scores between runners who received education on load management and underwent gait retraining based on step rate compared to those who only received education on load management [19]. A possible side effect of step rate-based gait retraining is calf muscle soreness. In one study, 43% of participants mentioned calf muscle soreness [49]. However, this did not affect running volume.

Tibial acceleration is a common variable of interest in gait retraining studies as a surrogate measure of loading rate. While a variety of techniques can be used to decrease tibial acceleration, some studies have found that runners are capable of lowering their tibial accelerations with visual feedback combined with instruction to land "softly" and "quietly" [14,39,41]. A study that used visual feedback of accelerometer data for gait retraining found significant reductions in tibial acceleration, VIP, VALR and VILR at post-training compared to pre-training [14]. Tibial acceleration, VIP, VALR and VILR all remained significantly reduced at 1-month, 6-month, and 12-month follow-ups.

Transitioning from a rearfoot strike to a forefoot strike has been shown to decrease loading rates while running [41]. Improvements in clinical outcomes that reach minimal clinically important difference in patients with PFPS that switch to a forefoot strike have been reported in a case series study [11]. These improvements remained at a 6-month follow-up. One randomized controlled trial found a significant reduction in pain in runners with PFPS that underwent gait retraining while the control group did not experience any significant changes in pain. However, subjects that underwent gait retraining reported calf soreness while undergoing gait retraining. Twenty-five percent of participants from the retraining group also reported ankle soreness at the 1-month follow-up when running more than 4 miles in a single session [42]. Chan et al. also found that runners who transitioned to a midfoot striking pattern did not display significantly different load rates compared to rearfoot strikers [43].

While fewer studies involving trunk lean were found, improvements in functional outcomes and decreases in pain were reported immediately upon the completion of gait retraining and at a 6-month follow-up [11]. Changes in functionality between pre-training and post-training reached minimal clinically important difference. These changes were similar to the changes seen in the treatment group that transitioned to a forefoot strike pattern.

#### **6. Clinical Application of Gait Retraining**

Sports injuries are complex and result from a combination of intrinsic and extrinsic factors [51]. Although biomechanics play a role in RRIs [27], it is important to highlight that a multifactorial perspective is required, and clinicians must identify the non-linear interactions between biomechanics and other aspects that may be related to injuries [51]. Gait retraining expands upon traditional approaches of addressing impairments in strength and flexibility to address abnormal biomechanics and motor control and should be considered as an aspect of a multimodal approach.

Most research on gait retraining strategies identified in this review (Table 2) focus on changing lower extremity mechanics to address knee pain. In PFPS, best evidence guidelines include exercises to improve hip and knee strength [52]. In the basics of the preliminary strengthening phase, the patient can also be considered in the cognitive stages of learning, using different exercises to gain intrinsic awareness of these muscles, and using visual cues to gain greater proprioceptive awareness. The treatment program then progresses to the associative stages of learning as tasks become more challenging. At this point, the participant should not only improve strength based on exercise prescription but improve ability to modify their alignment in a variety of tasks.

After addressing the pre-gait guidelines for management of PFPS, patients may further benefit by participating in gait retraining programs. Changes in pain and function up to six months were seen in runners with hip adduction greater than 20◦ that received visual feedback and were cued to "run with your knees apart with your kneecaps pointing straight ahead" [8,9], in runners with trunk flexion of 11◦ that were asked to "run with an increase in flexed trunk posture" [11], in runners with step rate between 160 to 170 steps/minute that were cued via an audio metronome set at 7.5–10% above their baseline step rate [11,19,49], and in runners that adopted a forefoot strike pattern using visual feedback and cues such as "run on your toes" and/or "run on the balls of your feet" [11,42].

The programs can be done using 8–12 sessions completed over 2–4 weeks (average of 2–3 gait retraining sessions per week). In this design, the feedback is provided continually in the first week and then is gradually removed in the second week. Run time progressively increases from 15 to 30 min [8,9,11,41,42]. This faded feedback program prevents dependency on external feedback and generates long-term retention [24]. Various other gait retraining programs have been proposed for patients with PFPS [19,49,53]. These include programs using five sessions over eight weeks [19], ten sessions over six weeks [49], and or only one session of ten minutes followed by four weeks of self-administration and monitoring increased step rate [53]. Each has shown improvements in symptoms and function and thus may be considered as an alternative to the above gait retraining prescription. Considerations for type of gait retraining intervention need to account for multiple factors including type of injury being treated, injury duration, time in season, and level of competition, with the goal of developing a long-term strategy to reduce risk for new or recurrent RRI.

Gait retraining implementation must consider the role of muscle strength and fatiguability, as different strategies will have different demands of the neuromotor system. An increase in trunk flexion is associated with greater peak hip extension moment [54], and transition to a forefoot strike is associated with greater peak ankle plantarflexion moment [55]. Strengthening the calf muscles may reduce the incidence of calf soreness that was reported for the step rate [49] and forefoot strike strategies [42]. Similarly, strengthening hip extensor muscles may facilitate a better transition to a gait with increased trunk lean [56]. Further, a combination of gait retraining strategies may be used to achieve the goals of the retraining program. Previous studies based on increasing step rate instructed patients to land softly [49] or to land softly and adopt a non-rearfoot strike pattern if necessary [19].

Regardless of the benefits for injured runners, very limited evidence supports the use of gait retraining for healthy runners. Athletes with VALR greater than 70 BW/s that received visual feedback and were instructed to "run softer" presented a 62% lower occurrence of RRIs in a year [57]. Only one retrospective study provided evidence that RFS runners present higher rates of prior RRIs than non-rearfoot strikers [58] and conversely another cross-sectional study found that a non-RFS was associated with calf muscle injuries and Achilles tendinopathy. Since no prospective studies have been performed, the transition to a forefoot strike in healthy runners cannot be uniformly recommended using evidence-based treatment [59,60]. Finally, limited evidence shows that the transition to a forefoot strike does not change running economy at fast speeds and limited-to-moderate evidence shows a decrease in running economy at low-medium speeds in recreational runners [59]. Therefore, changing the foot strike pattern to improve the running economy is not recommended. The potential of the other strategies to reduce the likelihood of RRIs and improve performance was not assessed.

While gait retraining has largely been studied for those with PFPS; limited work has been conducted on addressing biomechanical risk factors in non-PFPS RRI. Examples of gait retraining strategies that may be applied to runners presenting with various injuries are illustrated in Figures 1 and 2. It is important to note that gait retraining has only been used to treat runners with PFPS and chronic exertional compartment syndrome (CECS). Studies examining the effect of gait retraining on pain and functionality in many common running injuries do not currently exist, so specific recommendations for using gait retraining in injured runners with injuries other than PFPS and CECS cannot be made. Figures 1 and 2 serve only to provide examples on how a clinician may treat an injured runner using gait retraining based on literature surrounding gait retraining studies and risk factors for specific RRIs such as medial tibial stress syndrome [30,32], tibial stress fractures [61,62], iliotibial band syndrome [63,64], PFPS [28,29,32,42,65–67], CECS [46] and plantar fasciitis [33–35].

**Figure 1.** Examples of using gait retraining to treat bone related RRIs [22,30,32,61,62].

**Figure 2.** Examples of using gait retraining to treat other specific RRIs [28,29,32–35,42,46,63–67].

#### **7. Limitations of Current Gait Retraining Strategies**

Most studies characterizing gait retraining require participants to limit their running to gait retraining sessions in a laboratory setting [12,14]. This may discourage injured runners with an upcoming race, or those unwilling to take extended time away from running, from undergoing gait retraining. Studies that have allowed participants to continue with their own training schedules outside of gait retraining have found that runners can benefit from performing gait retraining while still participating in training outside of the laboratory [10,16,44]. Gait retraining completed outside of a laboratory setting while still continuing typical training has been observed to significantly reduce footstrike angle and increase step rate [50]. While this study did not report any negative outcomes for participants, RRI risk should be considered when allowing runners to continue with training while completing gait retraining, accounting for time for soft tissue and bone adaptations. Since this study only examined step rate manipulation, adjusting other variables may be less feasible outside of the clinical setting.

Despite the success found in gait retraining strategies employed in many of the studies examined in this review, a study conducted by Esculier et al. [19] found that education of proper training loads and education combined with gait retraining resulted in similar decreases in knee pain in runners with PFPS. Notably the session duration and frequency was lower than standard gait retraining programs and did not use a faded feedback design [8,9,11,14,18,41,42,47]. This discrepancy in results highlights that not all gait retraining protocols are equally effective, and that it is important to identify factors that may aid in the success of a gait retraining protocol.

Presently, it is unknown if gait retraining effectiveness is influenced by severity of injury. For example, improvement in pain and functionality in patients with PFPS following gait retraining only included participants that reported a pain level below a 7 out of 10 on a visual analog scale [42]. A separate investigation on gait retraining instructed PFPS patients to run only when their pain level was below a 2 out of 10 on the visual analog scale [19]. Thus, patients who report severe pain while running may benefit from undergoing other forms of treatment before beginning gait retraining.

There exists a clear gap in the literature surrounding gait retraining in terms of randomized controlled trials. Only two randomized controlled gait retraining studies have examined the effects of gait retraining as a form of injury treatment [19,42]. Both trials examined knee pain as an outcome measure and the two studies found contrasting results. While one study found a significant reduction in knee pain in the group that underwent gait retraining [42], the other study found there was no difference in knee pain in those who had undergone gait retraining and education compared to those who had underwent education alone [19]. Other studies using gait retraining did not record patient pain levels or did not include a control group. No randomized controlled trials have examined the effect of gait retraining on pain and functionality in RRIs other than PFPS. Further investigation into the effects of gait retraining as treatment for specific injuries may reveal that gait retraining is not equally effective for all types of RRIs. Identifying specific injuries that gait retraining is more effective in treating could increase the value of gait retraining as a rehabilitation tool for RRIs.

Only two studies included in this review reported adverse effects [42,49]. Both studies cited soreness of the calves that did not affect the ability of subjects to complete their training. One study reported ankle soreness at a 1-month follow-up after gait retraining. The time to achieve strength and tissue adaption must be individualized to reduce risk for RRI and gait retraining should be progressed gradually. More studies investigating potential adverse effects of gait retraining are needed before gait retraining can be fully recommended as a treatment strategy in injured runners.

#### **8. Conclusions**

While biomechanical risk factors are variable across RRIs, gait retraining may be used to modify potentially faulty running mechanics. An individualized and diagnosis-specific approach is important to address specific risk factors for the injured runner. Running mechanics can be modified using different forms of biofeedback and should use a faded feedback design for motor learning. Future studies, ideally in a randomized clinical study design, may clarify how different forms of gait retraining may be used, alone or in combination, to treat and prevent RRI.

**Author Contributions:** Conceptualization, L.W.G., M.M.B., A.S.T., L.W. and J.P.; formal analysis, L.W.G. and M.M.B.; writing—original draft preparation, L.W.G., M.M.B., J.P. and L.W.; writing review and editing, J.R.d.S.J., B.H., C.D.J., L.K.S., M.J.R., K.H. and A.S.T. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not applicable.

**Conflicts of Interest:** All authors have no disclosures related to this work. José Roberto de Souza Junior is a Fulbright visiting research student at the Spaulding National Running Center. Bryan Heiderscheit is a paid consultant to Biocore, has ownership interest in Science of Running Medicine LLC, is an advisory board member for Springbok Analytics and has received research funding (paid to institution) from the National Football League, National Basketball Association and GE Healthcare. Adam Tenforde serves as Senior editor for PM&R Journal. He gives professional talks such as grand rounds and medical conference plenary lectures and receives honoraria from conference organizers. He has participated in research funded by Arnold P. Gold Foundation (physician and patient care disparities), Football Player Health Study at Harvard (health in American-Style Football players), American Medical Society for Sports Medicine (bone density research), Uniform Health Service and Enovis (Achilles tendinopathy). He is a paid consultant for State Farm Insurance and Strava.

#### **References**


*Systematic Review*
