In terms of running, there is evidence that links mechanics with injury. This evidence provides the justification for altering these mechanics. Increased hip adduction and vertical impact loading have been most commonly associated with injury. More work is needed in order to understand the optimal way to retrain gait patterns in runners. The human body has a considerable ability to adapt. To provide individuals with the ability to alter faulty movement patterns in ways that can reduce injury risk is a powerful tool.
Key points
- •
Evidence links faulty running mechanics with injury and provides a justification and need for altering these mechanics.
- •
The human body has an amazing ability to adapt and leveraging this ability in order to reduce injury risk is very powerful.
- •
Altering habitual movement patterns requires practice and adherence to well-established motor control principles.
- •
More work is needed to determine optimal methods of retraining gait patterns.
Introduction
Why would it be necessary to alter someone’s natural running gait pattern? Running gait patterns are not self-optimized for each individual. Despite Nigg and colleagues’ theory of natural movement patterns, individuals often run with patterns that have been shown to be related to injury. Some patterns have also been associated with reductions in economy, such as increased vertical oscillation, increased stride length, and excessive arm motion. Although economy is important to overall running performance, this article focuses on the retraining of gait patterns with the goal of reducing the risk of running injuries. Because running injuries result from musculoskeletal overload, the issues contributing to this overload are reviewed first, with a focus on biomechanical factors. The inability of strengthening to alter running mechanics is discussed next. The important components of a retraining program are described, followed by a review of the literature on gait retraining to reduce injury risk in runners. The article concludes with a discussion of the future of conducting gait retraining in runners’ natural environments using wearable technology.
Introduction
Why would it be necessary to alter someone’s natural running gait pattern? Running gait patterns are not self-optimized for each individual. Despite Nigg and colleagues’ theory of natural movement patterns, individuals often run with patterns that have been shown to be related to injury. Some patterns have also been associated with reductions in economy, such as increased vertical oscillation, increased stride length, and excessive arm motion. Although economy is important to overall running performance, this article focuses on the retraining of gait patterns with the goal of reducing the risk of running injuries. Because running injuries result from musculoskeletal overload, the issues contributing to this overload are reviewed first, with a focus on biomechanical factors. The inability of strengthening to alter running mechanics is discussed next. The important components of a retraining program are described, followed by a review of the literature on gait retraining to reduce injury risk in runners. The article concludes with a discussion of the future of conducting gait retraining in runners’ natural environments using wearable technology.
Factors associated with running injuries
Running is a repetitive activity that typically occurs in a linear, forward direction resulting in a fairly invariant load with each foot strike. Because each foot strikes the ground approximately 625 steps per kilometer (1000 steps per mile), musculoskeletal tissues of the lower extremity become susceptible to cumulative overload, leading to overuse injuries. It can be assumed that every runner has a threshold for injury and that this threshold depends on several factors ( Fig. 1 ). The runner’s overall structure and static alignment plays an important role. For instance, foot structure has been implicated in different types of running injuries. As an example, high-arched runners have been shown to be prone to bony injuries, whereas low-arched runners are prone to soft tissue injuries. Although these structural factors need to be considered, they are effectively nonmodifiable. Dosage is another factor that can influence cumulative load. Training aspects such as how quickly the runner increases mileage, cross-training, and variation of running surfaces all contribute to the overall dosage of running. Dosage is clearly a modifiable risk factor, but there is a subset of runners who, despite reducing their mileage and resting appropriately, are unable to resolve their injuries, and these are the runners who most likely have an underlying mechanical factor that is not being addressed. This observation may explain in large part why the most common risk factor for a running injury is a previous injury. Mechanical factors can be divided into forces (kinetics) and movement patterns (kinematics). A runner showing abnormalities in either of these areas can experience excessive loading in their musculoskeletal systems. Runners experiencing both excessive forces and abnormal movement patterns are likely to have an even greater risk for injury.
The impact transient of the vertical ground reaction force is the immediate increase in force shortly after the foot contacts the ground ( Fig. 2 ). A sudden, large impact transient has been associated with injuries. In particular, greater vertical impact peak along with higher vertical average load rates (VALRs) and vertical instantaneous load rates (VILRs) have been linked to tibial stress fractures, plantar fasciitis, and patellofemoral pain. A recent prospective study suggests that having a greater VALR increases risk for developing a future running-related injury. Therefore, increased impacts might be considered to be a global indicator for injury, because the musculoskeletal system is composed of viscoelastic structures that are sensitive to rates of loading. It has long been established in animal studies of bone and cartilage that repetitive impulsive loads, even those within physiologic limits, can have damaging effects on musculoskeletal structures. As stated by these investigators, joint wear is determined not simply by the total force applied but by the degree and nature of the loading. In terms of movement patterns, there are some common malalignments that clinicians observe in injured runners. At the foot, these include increased stride length, excessive ankle dorsiflexion or inversion at foot strike, excessive foot eversion, and toe-in and toe-out gait ( Fig. 3 ). More proximally, increased genu varum and valgum, and anterior pelvic tilt and contralateral pelvic drop, are often seen. However, one of the most common abnormal movement patterns is the combination of excessive hip adduction, internal rotation, and pelvic drop ( Fig. 4 ). This particular malalignment has been associated with patellofemoral pain, iliotibial band syndrome, and tibial stress fractures in runners. Therefore, this malalignment pattern may be considered another global indicator for injury.
Strengthening alone is not enough
The most common therapeutic approach to altering faulty movement patterns is to strengthen the muscles that control that movement. For example, excessive hip adduction and internal rotation are often treated by strengthening gluteus medius (a hip abductor) and gluteus maximus (a hip external rotator). However, there is little support that strengthening these muscles, without neuromuscular retraining, translates into a change in movement patterns. Snyder and colleagues examined the effects of strengthening the hip abductors and external rotators on hip and knee mechanics during running in healthy active female runners. With the exception of a small increase in hip adduction excursion, the investigators noted no changes in hip and knee kinematics following a strengthening program compared with baseline. However, their study did not focus on individuals with abnormal mechanics, which may have limited its ability to examine potential changes in faulty mechanics. However, Mascal and colleagues reported on a case study involving a female runner with anterior knee pain associated with excessive hip adduction, internal rotation, and contralateral pelvic drop during a step-down maneuver. These abnormal mechanics were significantly reduced following a 14-week strengthening program. Note that the strengthening program focused on the proper mechanics during the single-leg stance and step-down exercises. Therefore, this additional neuromuscular reeducation could have influenced the participants’ movement patterns and alignment during this specific dynamic task. Willy and Davis examined the effect of a hip-strengthening program on otherwise healthy runners who showed excessive hip adduction with running. These individuals underwent a typical 6-week strengthening program addressing the hip abductors and external rotators. These runners showed an approximate 40% increase in strength of these muscles compared with a control group who did no strengthening. Despite this increase in strength, these runners showed nearly identical poststrengthening hip mechanics compared with their baseline values ( Fig. 5 ). Instruction and practice in proper single-leg stance activities were also included in the strengthening program. Like Mascal and colleagues, Willy and Davis noted significant improvements in the hip mechanics of the single-leg squat. However, these changes did not translate to alterations during the higher-demand activity of running. This finding clearly underscores the need for the retraining to be activity specific.
Brief history of gait retraining
The idea of altering gait patterns using feedback is not novel. The earliest forms of feedback were limb load monitors placed within the shoe of a patient. The aim of this type of feedback was to produce an equal load distribution between lower extremities during gait. Electromyography is one of the most widely used forms of feedback reported in the literature. Reports of improvements in gait symmetry in terms of spatiotemporal parameters and joint motion patterns have been reported with the use of electromyogram feedback. Feedback on joint angles has been provided through the use of electrogoniometers for patients with genu recurvatum. Almost all of these studies have reported successful results. However, none of these assessed the persistence or retention of these gait patterns over time, and most of these studies involved patients with neurologic disorders.
Reports of real-time feedback training then began to emerge in the orthopedic literature. In 2005, White and Lifeso provided real-time force feedback from an instrumented treadmill to patients who walked asymmetrically following a hip replacement. They reported a significant improvement in symmetry of ground reaction forces at weight acceptance following an 8-week (3 times per week) gait retraining program. In a related study, Dingwell and colleagues used an instrumented treadmill to improve the gait patterns of a group of unilateral, transtibial amputees. Before the training, asymmetries in the measured parameters were 4.6 times greater in the amputee group compared with the control group. These asymmetries were significantly reduced following the training. However, the long-term persistence of these changes was not monitored in either study and both investigations focused on walking gait. The physical demands of running are greater than those of walking, thus increasing the challenges of altering habitual running gait patterns.
Components of a retraining program
Altering any motor pattern that has become habituated over many years can be difficult, and this is especially true for retraining runners. Runners strike the ground approximately 625 steps per kilometer (1000 steps per mile). An individual who runs 32 km (20 miles) per week can log more than 1 million foot strikes per year, or 10 million foot strikes if they have been running for 10 years. Altering a motor pattern that has been reinforced over millions of cycles takes both guidance and practice. In a review on motor control principles, Winstein defined motor learning as a set of internal processes associated with practice or experience leading to a permanent change in the capability for responding. These processes are thought to be complex central nervous system phenomena whereby sensory and motor information is organized and integrated. This investigator suggests that learning a new motor program is enhanced with feedback provided in 2 phases. During the first, the acquisition phase, extrinsic feedback is provided on a prescribed schedule and helps to develop the connection between the extrinsic feedback (eg, real-time video of individual running) and the internal sensory cues (ie, proprioception) associated with the desired motor pattern. During the second, the transfer phase, the feedback is removed in a systematic fashion. The fading of this feedback prevents the reliance on it, and enhances the internalization, and thus learning, of the new motor pattern.
Although altering a motor pattern may be desired, it is also likely to alter the loads on the musculoskeletal system. For example, retraining a runner to become a forefoot striker in order to reduce vertical impacts reduces the load to the knee but increases the load to the calf. It is critical to fortify the calf musculature in order to reduce the risk of an overuse injury to this area. Therefore, it is important to include a specific strengthening program in any gait retraining intervention to anticipate increased demands to other components of the kinetic chain.
Altering movement patterns in runners
One of the earliest gait retraining studies in runners was conducted by Messier and Cirillo in 1989. Runners were seen 3 times per week for 5 weeks. Before each training session, subjects were shown a videotape of their running and were instructed on the features of gait they were to modify. These mechanics were subject specific and included characteristics such as excessive vertical oscillation, overstriding, excessive trunk lean, and excessive arm motion. This group of runners significantly altered the desired kinematic gait variables compared with a control group of runners who received no feedback before their training sessions. This study shows that runners are able to alter their mechanics with retraining. However, persistence of these changes following training was not examined, and it is therefore unclear whether true motor learning occurred.
With the development of real-time motion analysis systems, gait retraining could be augmented with real-time feedback on specific joint angles. Using this technology, Noehren and colleagues investigated 10 female runners with a history of patellofemoral pain (average duration of 75.7 months) who showed excessive hip adduction. These runners underwent 8 sessions (over 2 weeks) of gait retraining, learning how to activate the gluteal muscles in order to reduce hip adduction. Markers were placed on the lower extremities so that hip adduction could be monitored in real time. The hip adduction angle of the most involved side was displayed on a monitor in front of the runner during the stance phase of each foot strike. This angle was superimposed on a graph of a normal hip adduction trajectory with a shaded region denoting ±1 standard deviation of this mean value ( Fig. 6 ). Runners were asked to keep their hip adduction angles within the shaded region by modulating the activation of their gluteal muscles. The investigators were highly interested in the persistence of these changes. Therefore, they used a paradigm that incorporated the concepts of motor learning that were described by Winstein. Subjects gradually increased their run times from 10 to 30 minutes over the 8 sessions. During the first 4 sessions, runners were provided feedback 100% of the time (acquisition phase). During the last 4 sessions, feedback was faded (transfer phase) such that they received only 3 minutes of feedback during the last session (1 in the beginning, 1 in the middle, and 1 in the end) ( Fig. 7 ). Their gait data were examined at baseline, immediately posttraining, and at 1-month follow-up. Runners were able to reduce their hip adduction, internal rotation, and contralateral pelvic drop following gait retraining ( Fig. 8 ). In addition, they were able to maintain this at the 1-month follow-up. Controlling their hips better during running also led to reduced vertical load rates, which have been shown to be related to patellofemoral pain. Along with persistency, an indication of learning is the ability to transfer the new motor pattern to an untrained activity. These investigators noted that hip alignment also improved during the activity of a single-leg squat. Most important to the patients, there was a complete resolution of pain and significant improvement in overall function.
