Clinicians Guide to Ankle Dorsiflexion
Imagine running, squatting or going downstairs in a ski boot. Do you think your legs and pelvis would move differently? Have you ever tried to do those things with ski boots on? It’s pretty tough. What if you were allowed to loosen the boot a little? There would still be altered movement, but just not quite as bad. That is the nature of this post: Limited ankle dorsiflexion and its effects on kinematics and injury.
Previously, I made a post called The Definitive Guide to Pronation. This is another lengthy, detailed post, but now I’m moving up the kinetic chain. This post is on restricted ankle dorsiflexion (DF). It is really designed for clinicians, but if you are not a clinician you’re in luck. I created a “patient version” located here.
If you’re still reading, I’m assuming you have a working background education of anatomy and biomechanics.
There are a couple reasons that made me research this article: 1) The huge numbers of patients that I see with restricted ankle DF and 2) An open challenge by Dr. Greg Lehman in this post where he wrote, “Can you with certainty conclude that a lack of dorsiflexion is a true dysfunction? I think a massive post on restricted dorsiflexion and injury, form and performance would be cool. Any takers?”
Did he say “massive” post? Yes, but most readers will lose their attention span if I made it massive, so this isn’t massive. It’s just over 7,000 words. But, it’s still detailed and comprehensive. Let me warn you ahead of time, the research is plagued with problems. Various researchers have different definitions of what “limited” ankle DF is and how to measure it. Consistency is not a pillar of strength when it comes to this topic.
Because of the breadth of the post, I have made headings with hyperlinks within the post:
- What is “Normal” Ankle Dorsiflexion Range of Motion?
- How Does Reduced Ankle DF Alter Other Lower Extremity Kinematics?
- What Injuries are Correlated to Reduced Ankle DF?
- Can Limited Ankle DF Be Corrected?
- Conclusion: Is Limited Ankle DF a “True Dysfunction”?
Establishing normative data on range of motion (ROM) should be a simple task. I mean how hard can it be to reliably measure ankle DF? Well, very difficult when you consider the possibilities: Do you use a goniometer, inclinometer or tape measure? Open chain or closed chain? Knee straight, knee bent or both? Do you eliminate pronation or allow the subject to pronate? Healthcare is a funny business. We can make arguments out of the simplest things.
Without dragging out what should be the simplest variable in this whole article, I’ll get right to the point – there is no standardized method for measuring ankle DF, but the weight-bearing lunge position seems to be the most used, simplest and most reliable method for measuring ankle DF.
In non-weight bearing tests measuring ankle DF, the clinician is required to manually push the foot cephalad in order to get ankle DF. The amount of torque the clinician can generate is highly variable and to make things worse, they have to have a hand free to use a goniometer or inclinometer. You can see how this is a problem.
Conversely, in the weight bearing lunge test, the subject’s body weight provides more DF torque than any tester could, so we don’t have to worry too much about the tester not pushing hard enough. In addition, the clinician can have hands free to measure ROM.
Since pronation imparts dorsiflexion at the subtalar and midtarsal joints, another hurdle in measuring ankle DF ROM is limiting pronation. In other words, attempting to hold “subtalar neutral”. It has been shown that pronation can add as much as 8-10 degrees of dorsiflexion.
It is notoriously difficult to find and maintain subtalar neutral when measuring ankle DF, however, the weight bearing lunge test is a consistent, reliable method of measuring dorsiflexion. By making sure that the subject lunges forward with the thigh going straight ahead (see video below), pronation can be limited. If you don’t trust my word, try the test on yourself with the thigh moving straight ahead in the sagittal plane and measure then do the test while allowing yourself some femoral adduction and pronation. You will find your knee travels forward significantly more while in the femoral adducted and pronated position.
This technique is very reliable. Konor et al (2012) measured the distance from the toes to the wall with an intra-rater reliability (via intra-class correlation coefficient) of 0.99. By using an inclinometer or goniometer, the reliability was not quite as good. They did not specifically place the foot in a subtalar neutral position or utilize a small wedge placed under the medial aspect of the foot to maintain a more neutral position of the subtalar joint. Rather, they simply told the subjects to “progress his or her knee in an anterior direction.”
Bennell et al (1998) found an inter-rater reliability (via ICC) of 0.99 and again found the test less reliable when using an inclinometer
So, we have a very simple, extremely reliable test (inter and intra rater) that requires little to no equipment and is done weight-bearing. What else could you ask for? Well, validity is the answer to that question. We always come back to the question, “What are we actually measuring?” Since we are unable to totally eliminate subtalar and midfoot pronation we cannot be 100% accurate that we are measuring true talocrural dorsiflexion but it’s the best test we have. If you have read my lengthy blog post on pronation, you will realize that there is a healthy overlap of combined dorsiflexion and pronation at the talocrural and subtalar joints. The talocrural joint isn’t just a sagittal plane mover as we were taught in school.
Validity is going to be tough in any test designed to measure ankle DF. For example, Lundgren et al (2008) anchored pins in the bones of 5 subjects and measured the motion of the bones of the foot while walking. They found, “The ankle is often assumed to be the primary source of sagittal plane motion within the foot, but in four of five relevant subjects, the sagittal plane motion in the medial arch was greater than that at the tibio-talar joint.”
This and other studies like it prompted Gatt et al (2011) to state that when attempting to measure ankle DF, “the forefoot movement cannot be eliminated completely by placing the foot in any particular posture.” Thus, when validity is so difficult in any test measuring ankle DF, it is my opinion that the weight bearing lunge test is the easiest to implement with the most reliability of any other test. All of that being said, the Australian Physiotherapy Association’s Position Statement on Ankle Injuries still maintains that, “The weight bearing lunge test is a valid test for measuring ankle dorsiflexion range of movement”
The last component we need is normative values. Again, that is a difficult question and despite lengthy searches I haven’t come up with much. Bennell found ranges from 5-20 cm in the 13 healthy subjects they recorded, while Konor found a mean of 9.5 cm in the 20 healthy subjects in their study.
Due to the difficulty in finding normative values, I recruited the help of Dr. Craig Payne – a podiatrist and researcher with many publications, a University lecturer and owner of the website RunningResearchJunkie. He states that he typically looks for about 10cm but agrees that there is no established normative database. He stated that they have done some rough trigonometry based on the 10cm standard and came up with a tibial angle of 35-38 degrees.
Recently, I took an SFMA course where they use 5” (12.5 cm) as their pass/fail grade. Generally, I look for around 4-5 inches (10-12.5 cm). I try not to have a black and white line though where a patient with 9 cm is bad, but someone with 10 cm is good. There is certainly a gray area in there.
Finally, soft tissue extensibility from the gastrocnemius cannot be evaluated with the knee bent, so it is advisable to also check ankle DF with the knee straight. I would certainly advise checking ankle DF with the knee straight and knee bent. Both can be done in a weight bearing position. To measure ankle DF with the knee straight, simply perform a standing lunge test with the back leg straight, and measure the back leg’s ankle DF ROM. This will assist in differentiating gastrocnemius tightness from other sources of mobility restrictions.
I don’t have normative values on this. DiGiovanni et al (2002) tried to develop normative values for measuring ankle DF with the knee straight and came up with ≤5° (looking at people with symptomatic feet vs. a healthy control group). There are major caveats to that statement, however. Firstly, they did the test in a non-weight bearing position and with a goniometer. And yes, in their study they measured with more technical equipment, the inter-rater reliability in a regular clinical setting is terrible with both the non-weight bearing position (clinician’s ability to produce torque is questionable) and using a goniometer has poor reliability. Secondly, the reason they came up with ≤5° and not ≤10° is because they were able to accurately diagnose restricted ankle DF in the symptomatic group 76% of the time, whereas if they used ≤10° as a cutoff they were able to diagnose restricted ankle DF 88% of the time in the symptomatic group. They used the lower number because they were able to reliably avoid (in 94% of the cases) unnecessary treatment of those who were not at risk (the asymptomatic group).
Section 2. How Does Reduced Ankle DF Alter Other Lower Extremity Kinematics?
Proposed alterations to lower extremity kinematics include limited knee flexion, increased knee valgus, increased pronation, increased forward trunk lean (on squats), pelvic lordosis and early heel lift. Let’s look at each one individually:
2 i) Limited Knee Flexion:
Imagine lowering your body weight in a squat with ski boots on as I had suggested earlier. We would certainly see less knee flexion occur because the ankle dorsiflexion is limited. Since forward progression of the tibia is limited, more knee flexion would result in a posterior displacement of the body’s center of mass. Since the subject would fall backward at that point, knee flexion becomes limited.
Macrum et al. (2012) found that by having subjects stand on a 12° wedge (heel down) to simulate limited ankle DF ROM and perform a squat. Compared to normal conditions, subjects had a concomitant 15° decrease in knee flexion during the squat, representing a 16% decrease in knee flexion.
Fong et al. (2011) looked at 35 healthy subjects and measured their passive ankle DF ROM. They then had the subjects jump off a 30 cm box and measured lower extremity kinematics and other factors. They found that, “Greater passive ankle-dorsiflexion ROM was associated with greater knee-flexion displacement” In other words, limited ankle DF ROM was associated with reduced knee flexion during landing.
Conversely, DiStefano et al (2008) reported that peak knee flexion angle did not change when wearing an ankle brace to restrict ankle DF ROM during a drop landing test. However, in this study, the brace only restricted peak ankle DF ROM from 22° in the non-braced condition to 21° in the braced condition. Some may question whether limiting ankle DF by 1° is enough to cause measurable kinematic changes in the knee.
Complementing the reduced knee flexion during drop tests or squats is the fact that limited ankle DF is accompanied by knee hyperextension during the stance phase of normal walking gait. This was confirmed by Dudzinski et al (2013). Biomechanically, this makes sense, since forward progression of the tibia over the foot near the end of the stance phase requires ankle DF. If ankle DF ROM is not available, knee hyperextension would reduce the forward progression of the tibia. Therefore, clinicians should routinely check ankle DF ROM is patients with genu recurvatum.
2 ii) Increased knee valgus:
This is one of the more profound alterations in lower extremity kinematics due to limited ankle DF ROM. Not only in terms of implications, but in the amount of research confirming it.
Macrum et al. (2012) was the study listed above that had subjects stand on a 12° wedge (heel down) to simulate limited ankle DF ROM and perform a squat. Compared to normal conditions, subjects had a concomitant 18% increase in knee valgus.
Bell et al. (2012) studied 14 subjects and found that there was increased medial knee displacement during an overhead squat in those with limited ankle DF ROM.
Sigward et al (2008) studied 39 female soccer players and had them step off a 46 cm platform and land with both feet. They found that passive ankle DF ROM measurements negatively correlated with medial knee displacement. In other words, less ankle DF ROM = more medial knee deviation.
Unfortunately, this study like so many others, used a non-weight bearing method of measuring ankle DF with a goniometer. Not only inaccurate, I believe this would produce less of a negative correlation between limited ankle DF and medial knee displacement. The authors even state, “Given the nature of the task, the relationship between ankle range of motion measured on a weight-bearing position may have resulted in a stronger correlation with frontal plane knee excursion.”
Rabin and Kozol (2010) looked at 29 healthy females during a lateral step down test and found that limited ankle DF ROM was correlated with increased knee valgus as visually evaluated by 2 different physical therapists using visual inspection only.
Mauntel et al (2013) had 40 subjects perform a single leg squat and found that those with less passive ankle dorsiflexion also had greater medial knee displacement during the single leg squat. I found this study particularly interesting since the authors also looked at hip adductor and abductor activity during the task. The authors concluded that the hip activation patterns were because of the limited ankle DF.
Conversely, Bell et al (2008) found that during a squat, subjects with medial knee deviation during squat vs. subjects with normal knee frontal plane movement did not have statistically different ankle DF ROM’s. However, there was a trend towards limited ankle DF ROM and medial knee deviation – just not statistically different. The authors state, “Although this difference is not statistically significant, we believe the differences are clinically meaningful
and warrant further investigation.” In that same study, medial knee deviation was reduced by the use of a heel lift (i.e. reducing the need for ankle DF).
2 iii) Increased Pronation:
The idea of increasing your pronation to compensate for limited ankle DF is not hard to grasp. The video above is an easy way to test it out. Perform that test with the thigh travelling straight ahead and take the measurement. Then do it again, but this time, allow the knee to deviate medially to increase pronation. You will find it much easier to touch the wall. Conversely, try it with a small sock rolled up under the medial longitudinal arch to prevent midfoot pronation. Dorsiflexion will be more restricted in that scenario.
Karas et al (2002) explain it this way, “the DF available distally at the midtarsal joint, in conjunction with pronation, is used to supplement limited talocrural DF. As long as STJ pronation is possible, the midtarsal joint will provide additional functional dorsiflexion range during the progression of stance phase. Body weight will force the ankle foot complex into maximal STJ and midtarsal joint pronation, maximizing midtarsal joint dorsiflexion to supplement talocrural joint dorsiflexion.”
While this makes sense, I have only been able to find one clinical study that measured this distal compensation. Whitting et al (2011) had 48 men perform drop landings after their ankle DF ROM’s had been measured. They found that those with limited ankle DF, “displayed significantly more ankle eversion throughout most of the movement.”
In his book Human Locomotion, Michaud states that if ankle DF is limited, there will be compensatory subtalar and midtarsal pronation. He states, “This action tilts the oblique midtarsal joint axis to a more horizontal position, which allows the forefoot to dorsiflex more effectively about this axis.” This was confirmed in a cadaveric foot model by Blackman (2009) who found that simulated Achilles tendon contracture increased the severity of arch depression and forefoot abduction.
It is interesting to note, however that in a study by Cornwall (1999), they found that the magnitude of rearfoot eversion during walking was not greater in those with ankle DF ROM less than 10 degrees, however there was a delay in the timing of reinversion. However, this study solely looked at rearfoot motion and it appears that much of the compensatory motion associated with limited ankle DF occurs in the midfoot and forefoot.
2 iv) Increased Trunk Lean:
In order to squat with the thighs parallel to the floor, approximately 22° of ankle dorsiflexion is required.
By limiting the forward progression of the tibia over the foot, restricted ankle DF ROM will limit knee flexion and create a posterior shift of the body’s center of mass. This will cause compensatory increased trunk lean in order to prevent the subject from falling backward. In order to increase the forward trunk lean, you need to increase hip flexion. This increases the torque on the hips.
Fry et al (2003) reported this in their study where they limited the forward progression of the tibia by having subjects squat with a board placed vertically from their toes upward (see pic below)
By increasing the forward trunk lean, there was a concomitant increase in hip flexion and so torques were increased at the hip and the low back. The authors concluded that by restricting the forward progression of the tibia, “it is likely that use of greater relative loads for a restricted squat could produce excessive forces at the hips and low back”
In her very popular book “Gait Analysis”, Perry reports that limited ankle dorsiflexion (she calls it excessive plantar flexion, but states that the two terms are interchangeable when discussing gait) will have three main proximal compensations: 1) Early heel lift 2) Knee hyperextension and 3) “Forward lean of the trunk with anterior tilt of the pelvis”
2 v) Increased Ground Reaction Force
Yes, I know I said I was going to list the kinematic variations, and now I’m getting into ground reaction force (GRF). In the next section of the blog, I talk about injuries associated with limited ankle DF and a few of them have to do with increased GRF, so I thought I’d include this.
I don’t think I have to go into a long dissertation about this because biomechanically, it makes sense – if the ankle DF excursion is limited, and knee flexion excursion is limited, you are going to have to absorb landing impact over a shorter distance. In other words, a “stiffer” landing. Therefore, GRF should increase.
This was confirmed in a study by Fong et al (2011) where they found increased ground reaction forces in individuals who had less ankle DF ROM.
2 vi) Impaired Static and Dynamic Balance Testing
Mecagni et al (2000) performed ankle DF ROM testing and then some dynamic and static balance testing. They found significant correlations between reduced ankle DF and poor balance. There was less correlation between ankle DF and static balance than there was with dynamic balance testing and the gait parameters in the POMA balance testing procedures.
There are two studies that have looked at the Star Balance Excursion Test (SBET) – Hoch et al (2012) and Basnet et al (2013). Both found significant associations between reduced ankle DF and SBET test scores. Obviously, the worst scores were for the forward reach test (the weight bearing leg needs a lot of ankle DF ROM to score well) but there were also correlations for the posterolateral test and composite SBET scores.
Section 3. What Injuries are Correlated to Reduced Ankle DF?
The previous chapter has outlined the most commonly documented kinematic changes that occur as a result of limited ankle DF. Those compensatory changes will alter loads and possibly create injury. Here is a list of the most well documented injuries associated with limited ankle DF
3 i) Patellar Tendinopathy
There are two significant studies that have looked at how limited ankle DF can be associated with patellar tendinopathy.
Backman and Danielson (2011) performed a prospective study on 90 junior elite basketball players. They measured a number of potential risk factors for patellar tendinopathy. They found that those with limited ankle DF were at a significantly higher risk of developing patellar tendinopathy. More specifically, if the tibial angle was less than 36.5°, the basketball players had a risk of 18.5% of developing patellar tendinopathy in their dominant limb and 29.4% in their non-dominant limb. Those with a tibial angle of greater than
36.5° had a risk of 1.8% of developing patellar tendinopathy in their dominant limb and 2.1% in their non-dominant limb. So basically, they found there was a 10 fold higher risk (more on this in section 5) of patellar tendinopathy in their ankle DF was limited to a tibial angle of less than 36.5°. This angle corresponds to what Dr. Craig Payne had suggested to me and I had reported earlier in this post.
Mallarias et al (2006) looked at 113 female volleyball players to see the association between patellar tendinopathy and various performance factors including sit and reach flexibility, ankle dorsiflexion range, jump height, ankle plantarflexor strength, years of volleyball competition and activity level. In the end, limit ankle DF was the only factor associated with patellar tendinopathy.
3 ii) Ankle Sprains
Here is the problem with ankle sprains – if you have limited ankle DF, you may be more likely to suffer from ankle sprains. However, one of the biggest manifestations of an ankle sprain is that your ankle DF ROM becomes limited. It’s a bit of a snowballing effect, which may contribute to why recurrence rates are so high
Pope et al (1998) did a prospective study on 1093 army recruits over a 12 week intensive training program. They looked at ankle DF ROM and they tracked only 5 injuries: ankle sprains, stress fractures of the foot or tibia, tibial periostitis, anterior compartment syndrome and Achilles tendonitis. The mean ankle DF ROM was 45 degrees as measured by the weight bearing lunge test. Those with an ankle DF ROM of 34 degrees (the lowest DF ROM group in their study were 2.5 times more likely to suffer from one of those 5 injury types. Some injuries were higher. For example, stress fractures had no correlation, but ankle sprains were 5X higher in those with limited ankle DF. The results showing no correlation to stress fractures is in stark contrast to Hughes (1985) who found a 4.6X increase in metatarsal stress fractures in those with limited ankle DF ROM.
The reason I listed the paper by Pope is because it was a prospective study. Willems et al (2005) is another example of a prospective study showing that limited ankle DF ROM increases risk of ankle sprains. The relationship between ankle sprains and limited ankle DF has been documented in other research papers that weren’t prospective: Hoch (2012), Yang (2002), Drewes (2009) (measured during jogging)
3 iii) Plantar Fasciitis(osis)
If it is true that limited ankle DF ROM is compensated by a tilt in the midfoot axis of rotation to a more horizontal position in order to gain dorsiflexion from the midfoot, it would make sense that there would be an increase in tensile forces in the plantar fascia. To measure this, we need to put a strain gauge in your plantar fascia. Any volunteers?
Cheung et al (2006) performed a 3D modeling of this idea which confirmed that “Increasing tension on the Achilles tendon is coupled with an increasing strain on the plantar fascia.” This is interesting, but certainly doesn’t prove that limited ankle DF increases risk of plantar fasciitis.
When reading all the “experts” online, you would think that it was a given that there was a strong relationship between limited ankle DF and plantar fasciitis, but it’s not that clear. Rome et al (2001) found no relationship at all, while Irving et al (2007) found in 80 subjects suffering from chronic heel pain compared to 80 healthy subjects, there were more people with excessive ankle DF in the chronic heel pain group compared to the healthy group (33% vs 19%). The mean ankle DF angle for the heel pain group was 45 degrees and 40 for the healthy group.
Those studies are in direct contrast to other studies that have found a relationship between limited ankle DF and plantar fasciitis. For example, using 10° of ankle DF with the knee extended as a cutoff point, Bolivar et al (2013) found that limited ankle DF presented a sensitivity of 100% and specificity of 96% for predicting plantar fasciitis for the participants in this study. Kibler et al (1991) found that ankle DF was limited in 37 of 43 feet affected with plantar fasciitis. Additionally, Patel and DiGiovanni (2011) found that in 254 patients with plantar fasciitis, 83% of them had restricted ankle DF ROM. Riddel et al (2007) looked at 50 patients with plantar fasciitis vs. 100 controls and found, “The risk of plantar fasciitis increases as the range of ankle dorsiflexion decreases.”
3 iv) Miscellaneous Injuries Grouped Together
Trust me , I know this sub-heading has a strange title, but I didn’t know what to call it. There are a number of studies that have found limited ankle DF is associated with higher injury rates in general.
DiGiovanni et al (2002) looked at subjects with “foot symptoms” vs a healthy control group. They found that ankle DF was limited (≤5°) with the knee straight (65% of the symptomatic group vs 24% of the control group) and was also limited (≤10°) with the knee bent (29% of the symptomatic group vs 15% of the control group.
The idea of foot injuries increasing due to limited ankle DF is not a novel idea. Hughes (1985) found soldiers with limited ankle DF were 4.6X more likely to sustain metatarsal stress fractures.
Gabbe et al (2004) had 126 Australian Rules Footballers (community level players) undergo a “battery of musculoskeletal screening tests” at the beginning of the season and tracked who got injured. Restricted ankle DF as measured by the weight bearing lunge test was the ONLY test with a significant association to who got injured. I have been unable to attain the full text, so I don’t know what the ROM cutoff for “normal” was and I don’t know what the other tests in the “battery of musculoskeletal screening tests” were.
Nitz and Choy (2004) measured ankle DF ROM with the weight bearing lunge test on 372 women aged 40-80. They looked at the number of falls over a 12 month period and found that the non-fallers had around 8° more ankle DF than those who fell at least twice. These results were irrespective of age. Also, they found no association between the number of falls and activity level.
Tibrizi et al (2000) looked at the ankle DF ROM on the uninjured side of 82 children with lower extremity injuries and compared the values with a control group of 85 children with upper extremity injuries. They found that there was a significant difference in ankle DF with the knee extended (5.7° vs 12.8° in control group) and also with the knee flexed (11.2° vs 21.5° in control group). Obviously this study has flaws in its research methods, but the authors site other studies that ankle DF ROM is generally equal bilaterally, so they maintain that measuring the uninjured side is valid. They concluded, “Our observations suggest that children with injuries to the ankle have less inherent flexibility before the injury. We believe that this contributes to the cause of the injury.”
3 v) The Rabbit Hole
“You take the blue pill, the story ends, you wake up in your bed and believe whatever you want to believe. You take the red pill, you stay in Wonderland, and I show you how deep the rabbit hole goes.“
Given the number of biomechanical changes that limited ankle DF causes, we can start extrapolating other injuries that theoretically could result. It’s the “If This, Then That” argument. For example…
- Non-contact ACL Injuries: Given that limited ankle DF causes a) increased GRF during landing, b) reduced knee flexion during landing and c) knee valgus when landing, can we say limited ankle DF increases your risk of ACL tears? Well, it hasn’t been studied, but those biomechanical conditions that I just listed are well established to be three of the most common risk factors for ACL tears Hewett et al (2005), Yu et al (2006), Griffen et al (2005).
- Patellofemoral Pain Syndrome: Given that limited ankle DF causes a) reduced knee flexion during landing and b) knee valgus when landing, can we say limited ankle DF increases your risk of PFPS? Both increased internal femoral rotation and decreased knee flexion during landing were risk factors for PFPS as identified by Boling et al (2009). Knee valgus has been identified as a risk factor for PFPS in a number of other studies including Waryasz (2008). There are some contradictory studies, however they are usually done on subjects currently suffering from PFPS, so the subjects may have been compensating for the pain. In fact, Leitch et al (2012) did look at ankle DF in runners with a history of PFPS (not currently suffering from PFPS) and found reduced ankle DF; however this study was retrospective, so cause-effect is difficult to establish.
In a recent interview with Lower Extremity Review, researchers Chris Powers and Darin Padua talked about the similarities between ACL injury and PFPS. Powers stated, “I wouldn’t be surprised if at some point we figure out that patellofemoral pain is a predictor of who goes on to tear their ACL.”
Well… it depends.
A specific intervention may help some people with restricted ankle DF but that same intervention may not help others or even make them worse off. There are many causes of limited ankle DF. It could be a soft tissue extensibility problem (gastrocnemius, soleus or potentially, but rarely tibialis posterior or peroneals), it could be an osseus block from a congenital anomaly, a tight posterior joint capsule/posterior tibiotalar ligament/posterior talofibular ligament, anterior tibiotalar exostosis or from restrictions in the midfoot/forefoot.
When determining whether the cause is soft tissue or joint based, I will leave this up to the clinicians. Obviously some causes are easy to determine (knee flexed vs. straight for gastroc) and other more difficult that require skilled palpation or advanced imaging. When discussing osseous congenital anomalies in his book Human Locomotion, Michaud states, “The most common deformity affecting ankle dorsiflexion is the flattened talar trochlea.” He goes on to states that, “Another bony anomaly that may restrict ankle dorsiflexion relates to a congenitally wide anterior talar dome.“
Another possibility for bony restriction could be an anterior ankle impingement exostosis. If you try and force ankle dorsiflexion on a condition like that, you are very likely to create more problems. Alternatively, if there is any osteochondral lesion on the anterior aspect of the talar dome and you try and force dorsiflexion, you are certainly asking for trouble. This is yet another reason why my pet peeve lately is internet “experts” who make blog posts about how to fix or prevent injuries. Without proper examination, sometimes more harm than good is done.
4 i) Increasing Dorsiflexion Following an Ankle Sprain:
Following ankle sprains, restricted ankle DF is a common outcome. It is important to improve the ankle DF ROM because limited ankle DF is a risk factor for future ankle sprains. In a systematic review of the literature by Terada et al (2013), the authors concluded that, “A static-stretching intervention as part of a standardized home exercise program had the strongest effects on ankle dorsiflexion improvement after acute ankle sprains.” That is not to discount the idea of passive, manual mobilizations in the form of Maitland or Mulligan mobilization techniques. The authors acknowledged the idea that, “posterior gliding of the talus may be restricted during dorsiflexion because disruption of the anterior talofibular ligament may induce anterior subluxation and internal rotation of the talus on the mortise and anterior and inferior displacement of the distal fibula.” However, the authors fell short of endorsing manual mobilization techniques and stated, “the clinical relevance of conclusions drawn from the current literature is limited because the associated effect sizes were small to moderate.”
4 ii) Increasing Dorsiflexion in Non-Injured Ankles:
With respect to stretching to increase ankle DF ROM in non-injured adults, it’s a mixed bag.
Radford et al (2006) performed a systematic review of the literature and were able to find 5 RTC’s. They found that there was a benefit to static stretching, although whether the gains seen are clinically significant is questionable. For example, they found that subjects who stretched <15 minutes of resulted in a 2.07° increase whereas 15–30 minutes of stretching resulted in a 3.03° increase and stretching for >30 minutes resulted in a 2.49° increase. Is 3° of gain going to change your kinematics or lessen the risk of injury? We don’t know.
Here is the problem with the Radford study and all others comparing calf stretching: There is no consistency! Radford et al did a fantastic job with what they had, but they’re trying to pool data where some studies had their subjects stretch during weight bearing, others didn’t. Some stretched with the knee straight, others with the knee bent. Some used pulleys and weights to assist in stretching, others didn’t. Some controlled for pronation, others didn’t, some stretched more days a week than others, and some held stretches longer than others. In order to measure gains from the stretching protocols, some studies measured weight bearing, others with goniometers, some with knee straight, others with knee bent….AGH!
Since the Radford paper in 2006, there a couple studies that have emerged: Johnson et al (2007) had 20 elderly women stretch 5 days a week for 6 weeks and found a mean increase in ankle DF ROM of 12.5°. Also, Whitting et al (2011) had 48 men perform drop landings after their ankle DF ROM’s had been measured. There were a number of different interventions in this study and I am not going to attempt to explain them all. Suffice it to say that following 6 weeks of static stretching for 5X 30 second hold stretches for 5 days/week for 6 weeks, ankle DF ROM’s increased 3.1° (7%) with the knee flexed and 5.7° (15%) with the knee extended. The group that stretched not only attained an increase in ankle DF ROM, they also witnessed a decrease in peak Achilles tendon force during their drop landings as well as using a significantly reduced percentage of their peak ankle eversion and ankle DF ROM capacity. They did not, however, reduce their peak GRF. Still the authors concluded that by utilizing static stretching to increase their passive ankle DF ROM, the subjects attained “protective adaptations in terms of the injury potential postulated by the mechanisms observed during pre-intervention baseline
Another method used to increase ROM’s is to perform eccentric training on the muscles that cross the joint. This method has been used successfully on the hamstrings (Nelson et al 2004) where eccentric training was shown to work just as well as static stretching over a 6-week program. The posterior calf musculature is no different: Mahieu et al (2008) had their subjects undergo a 6 week eccentric program and found a significant increase in ankle DF ROM. In addition, Whitting et al (2011) also found increases in ankle DF ROM following an eccentric program. Without getting sidetracked too much, this is an appropriate time to talk about ankle DF stiffness, as opposed to ROM. Stiffness is measured via a dynamometer. In both the Whitting and the Mahieu studies, ROM’s were increased, but stiffness was not changed. I am hesitant to advise clinicians to start testing ankle DF stiffness since more equipment is required and in a completely separate paper by Whitting et al (2013) is was found that ankle DF stiffness is poorly correlated to ankle DF ROM.
Yet another method of increasing ankle DF is with night splints, but studies report very poor patient compliance with them, plus you can’t stretch the gastrocnemius without putting the patient in a splint that also travels past the knee to keep it in extension.
Still another method to increase DF is through joint manipulation, Mulligan MWM, Maitland mobilizations. Personally, being an instructor for Active Release Techniques (ART) I’m partial to ART on the posterior tibiotalar and talofibular ligaments. I have found fantastic success with this technique; however I can’t point to any research backing it up. Being a clinician is often a blend of what we have learned from clinical experience and what we learn from reading studies. If RTC’s were our only source of treatment guidance, we’d all be in trouble.
That being said, there are a few studies on manual therapy and increasing ankle DF in non-injured subjects. Some have been negative – Nield et al (1993), Fryer et al (2002). While some have been positive – Vicenzino et al (2001), Dananberg et al (2000), Guo et al (2006). The Guo paper was particularly interesting as not only did they find increases in ankle DF ROM following Mulligan mobilization, but they ROM’s were still significantly different after 2 days of follow up. Additionally, they measured gait parameters and found increases in step length during slow walking and interestingly, increases in velocity. On the other hand, Craib et al (1998) found improved running economy was associated with decreased ankle DF.
As a side note, I have been reading the Twitter world lately and it seems there are a lot of therapists jumping on the “manual therapy does nothing mechanical. Any changes seen are due to descending inhibition and placebo.” bandwagon. For what it’s worth, Collins et al (2004) performed a study on Mulligan mobilization with movement on subjects with subacute ankle sprains and concluded, “Results indicate that the MWM treatment for ankle dorsiflexion has a mechanical rather than hypoalgesic effect in subacute ankle sprains.” I have provided the link to the full text if you care to read it.
Mike Reinhold has provided some great examples on self-mobilization techniques for increasing ankle DF.
4 iii) The Unfixables:
In cases of congenital osseus blocking of dorsiflexion, compensations need to be made. Theoretically the addition of a heel lift will place the ankle in a more plantarflexed position which should add some dorsiflexion excursion to any tasks requiring ankle DF. The changes at the ankle should be seen up the kinetic chain. This technique has been shown to help in the study by Bell et al. (2012) where they found that subjects displaying medial knee deviation during squat also had 20% less ankle DF ROM compared to subjects that did not show medial knee deviation during a squat. By subsequently using a heel lift, subjects reduced medial knee deviation during a squat.
This post has covered both proximal and distal compensations to limited ankle DF. The big distal compensation is pronation to allow greater dorsiflexion distal to the talocrural joint. Remember that Lundgren et al (2008) found that “the sagittal plane motion in the medial arch was greater than that at the tibio-talar joint.” If such compensation takes place and this person goes into a running shoe store, the first thing they’ll hear is “You overpronate, we need to stop that” and the customer will get off the shelf orthotics and/or pronation control shoes. So now, the distal compensation has been reduced, leaving the person’s body with a choice: Add more strain back on to the ankle joint and potentially injure that joint, or compensate further up the kinetic chain. If that person then gets patellar tendinopathy or other knee or hip injury, who’s fault is it? I would argue the person who put them in the device that took away the distal compensation.
Section 5. Is Limited Ankle Dorsiflexion a True Dysfunction?
At the beginning of this post, I referenced another blog post by Dr. Greg Lehman. He writes, “athletes can get by without restricted dorsiflexion in many sports. Do we always want to go changing this? Can you with certainty conclude that a lack of dorsiflexion is a true dysfunction?”
Well I think I can say, after reviewing all of this literature, that limited ankle DF certainly has the capacity to increase knee valgus, decrease knee flexion at the bottom of a drop jump, increase pronation, and change dynamic balance. Ostensibly due to the kinematic changes, the risk of certain injuries increases including patellar tendinopathy, ankle sprains and lower extremity injuries in general. These have been seen in many prospective studies such as Backman and Danielson (2011), Mallarias et al (2006), Pope et al (1998), Willems et al (2005), Gabbe et al (2004). This isn’t including the potential risk factors by extension that I covered in the section “The Rabbit Hole”.
There are very few unconditional and absolute statements anyone can make about healthcare. Just about every answer could start with two words: “it depends”
When we talk about risk, we need to differentiate between absolute risk and relative risk. For example, in the Backman and Danielson (2011) paper, they found that 18.5% of basketball players developed patellar tendinopathy in their dominant limb if their ankle DF was less than 36.5°. If their tibial angle was greater than 36.5°, only 1.8% of them developed patellar tendinopathy in their dominant limb. Therefore, since the risk was 10X more, it is easy to say there is a 10 fold increased risk. However this is only the relative risk. The absolute risk is really only 17 players out of 100 will get patellar tendinopathy because of limited ankle DF.
Alternatively, we could also say 81.5 out of 100 basketball players WILL NOT develop patellar tendinopathy if they have limited ankle DF and 98 out of 100 players WILL NOT develop patellar tendinopathy if they have great ankle DF (read more on absolute vs relative risk here). When you say that 81 out of 100 players with limited ankle DF won’t develop patellar tendinopathy, that doesn’t sound too bad!
This plays into the statement “athletes can get by without restricted dorsiflexion in many sports.” This is absolutely true. However, risk is risk and if I can reduce the risk of getting injured in my patients, I will certainly tell them. Let’s get into the world of analogies: Imagine some guy, Joe, likes to go to a bar every night. Let’s say he has a 20% chance of getting in accident when he drives home from the bar if he’s been drinking. Conversely, he has a 2% chance of getting in an accident if he doesn’t drink. In other words, a 10X increase of getting in an accident if he drinks. However, if he drinks before he drives, 8 out of 10 times he’s OK. One day, he comes to you and asks why he got in an accident (aka patellar tendinopathy). Do you tell him not to drink and drive (aka increase his ankle DF)? I would argue that I would try and get him to stop despite the fact that 8 out of every 10 times he was OK.
Similarly, not all athletes with limited ankle DF will have an injury because of it, but I wouldn’t want to be playing sports or be a runner with increased risk of injury. Some people discount the idea of movement assessments, but ankle DF is one movement assessment I will continue to utilize on most patients and if it is significantly limited I will suggest that they try and correct the dysfunction.
Unfortunately, with the lack of consistency in the research of what “limited” ankle DF is defined as and how to measure it, we can’t make blanket statements. In addition, due to the many causes of limited ankle DF, the success of attempting to restore mobility can vary greatly. Still, I believe that we have enough data to say that limited ankle DF causes biomechanical compensations and increases the risk for various injuries.