July 2014

Revisiting the status quo for AFO strap placement

afo-mainConventional wisdom says dorsal straps for ankle foot orthoses should be positioned at a certain angle to resist heel rise. Now preliminary research from Georgia Tech is calling that practice into question, but not all clinicians are convinced that the findings warrant a different approach.

By P.K. Daniel

Practitioners worldwide for years have positioned dorsal straps for ankle foot orthoses at a 45° angle to the bony structure, even though there was no research to support this practice. Now researchers from the Georgia Institute of Technology in Atlanta are initiating a discussion about whether that angle is actually the ideal placement from a biomechanical perspective.

There seems to be a tendency among some practitioners to maintain the status quo, simply because it’s been considered the “best practice.”

“The role of an orthotist has been built on best practice in a lot of areas,” said Carey Jinright, LO, MSM, owner of Precision Medical Solutions in Auburn, AL. “This strap placement is one of those practices that have been used for many years—forty five degrees across the instep of the foot just distal to the talus is the typical point and angle of strap placement. The goal of the strap is to push the heel back into the heel cup within the molded footplate.”

But, a few years ago, Georgia Tech researchers decided to challenge the status quo. They questioned whether 45° really is the ideal placement for resisting heel rise, which can cause blistering and ulcers.


Pilot study

Tyler Manee, CPO, is a prosthetist-orthotist at Ability Prosthetics & Orthotics in Frederick, MD. For his thesis in Georgia Tech’s Master of Science in Prosthetics and Orthotics Program, he led a pilot study in 2011 on dorsal strap placement —a study that concluded 45° was not, in fact, the optimal placement. To minimize foot movement and maximize comfort through lower corrective forces, a dorsal control strap should be located as proximal as possible and the force application point should be angled 90° to the bony structure, Manee and colleagues concluded.1

Figure 1. Movement of the foot is restricted by means of a force couple, where the dorsal strap serves as the corrective force and is opposed by two counter forces, one at the posterior proximal region of an AFO’s calf section and a second at the plantar aspect of the metatarsal heads applied by the foot section of the orthosis. (Image courtesy of Géza F. Kogler, PhD, CO.)

Figure 1. Movement of the foot is restricted by means of a force couple, where the dorsal strap serves as the corrective force and is opposed by two counter forces, one at the posterior proximal region of an AFO’s calf section and a second at the plantar aspect of the metatarsal heads applied by the foot section of the orthosis. (Image courtesy of Géza F. Kogler, PhD, CO.)

Straps were positioned at 80°, 90°, and 100° from the dorsal line on Manee’s cadaveric model, which was designed to mimic heel rise within the AFO and not necessarily the heel rise phase of gait. Therefore, it’s possible the results don’t transfer to a walking scenario, Manee said.

Although Georgia Tech researcher Géza F. Kogler, PhD, CO, who worked with Manee on his initial study, said the tendency for the heel to rise is greatest during the swing phase of gait, Thomas V. Dibello, BS, CO, currently in private practice at Dynamic Orthotics and Prosthetics in Houston, TX, where he is the owner and chief executive officer, said this can occur at any phase of gait.

The goal, Dibello said, in any lower limb orthosis is to keep the patient’s heel seated in the heel of the device.

“The angles that this study describes maximize the lever arms that apply the stabilizing force to the foot to hold it in the orthosis,” Dibello said. “Those lever arms do not change irrespective of the phase of gait that you may be in.”

A closer look

Kogler, who is the director of the Clinical Biomechanics Laboratory in the School of Applied Physiology at Georgia Tech, has expanded on Manee’s pilot study to further determine the optimal angle and longitudinal axis placement to minimize pistoning within an orthotic device. The results, he concluded, were about the same.

To avoid foot motion, Kogler said, clinicians should aim for the most proximal placement (longest lever arm) at an obtuse angle (105°) with the controlling force perpendicular to the bony structure. Prosthetics & Orthotics International epublished the Tech group’s most recent findings on June 11.2

In the expanded in vitro cadaveric study, researchers used nine test conditions to quantify the force of a dorsal control mechanism. They used three angles (75°, 90°, 105°) and three longitudinal axes at 2-cm increments.

The longest lever arm (proximal location) applied at an obtuse angle (105°) required the least amount of force (55.6 N) to limit foot motion, whereas the shortest lever arm (distal location) at the acute angle (75°) required the greatest force (90.4 N) to limit foot motion. Although the talus served as the proximal longitudinal axis position, Kogler said it was important that clinicians be aware that the foot’s dorsal surface does not necessarily represent the orientation of the underlying bony structure. The soft tissue structures traversing the foot may impede optimal control of heel rise within an orthotic device, so practitioners should consider a clinical assessment of the anatomy at the location of the orthotic constraining force (ie, the dorsal foot strap).

“It’s surprising to me after all of these years that no one has looked at exactly the real details of the optimal way that force can be applied,” Kogler said. “And there are many questions that need to be asked, like: How can that force be applied comfortably? And how can that force be applied with the least amount of force yet still get the job done of holding the ankle at a ninety-degree position?”

Clinical implications

Instead, with no scientific studies to refer to, orthotists have used their experiences to determine how to improve fit, comfort, and function related to orthotic control of heel rise in lower limb orthoses. The clinical relevance of the Georgia Tech research, Kogler said, is its finding that a dorsal foot control strap with proper placement and angular orientation can provide maximum function and comfort to users.

Figure 2. Practitioners traditionally position the dorsal foot strap at 45° with refer- ence to the plantar aspect of the foot section of the orthosis. The Georgia Tech data suggest this customary positioning isn’t optimal since the resultant angle appears to be acute relative to the dorsal surface and the bony structure of the foot. (Image courtesy of Géza F. Kogler, PhD, CO.)

Figure 2. Practitioners traditionally position the dorsal foot strap at 45° with refer- ence to the plantar aspect of the foot section of the orthosis. The Georgia Tech data suggest this customary positioning isn’t optimal since the resultant angle appears to be acute relative to the dorsal surface and the bony structure of the foot. (Image courtesy of Géza F. Kogler, PhD, CO.)

Some practitioners, like Dibello, stressed that studies such as this one are vital in developing evidence to support improved practices. The implications of initial studies are not always immediately apparent or applicable to what is happening in clinical practice, however, and some practitioners questioned aspects of the study, including the use of cadavers.

“With all of the digital technology and electrode placement advancements that we have today, I do not understand why this was carried out using a cadaver,” Jinright said.

The study authors explained using cadavers was necessary to isolate the foot movement, and that in research, investigators have to control variables to gain new knowledge.

“Our project was focused on strap location for controlling heel rise within the AFO,” Manee said. “Plenty of other research would be required to determine optimal strap placement, period. We’re laying a foundation, and for a foundation you want to eliminate as many variables as possible. And then add them in slowly.”

Another test aspect that practitioners questioned was the cylindrical aluminum bar that was used in the mechanical testing apparatus instead of an actual strap, which could potentially distribute the forces differently in a clinical setting. The researchers explained the bar needed to be rounded for their experiment because, as they turned the force applicator to test the various angles, the point of application needed to remain the same.

“A strap has edges and rectangular geometry laying over a curved fleshy anatomy, and the force distribution over the width of the strap cannot be known except that it is extremely [improbable that the force is] uniform,” Manee said. “I would love to do future tests to find a way to use a strap, but this experiment was focused on laying a baseline proof of concept for the theory.”

Dibello said using a nondeflectable material makes sense at this point, in this type of analysis, as it eliminates a variable.

Both the authors and practitioners acknowledge further research is required and that this study was designed to lay the groundwork for further investigation.

“I would encourage the author to advance the study to human subjects, and then eventually various strap material configurations,” Dibello said.

Another practitioner also suggested using an AFO on people.

“I think they could have used a different model to achieve real-life inside AFO measurements,” said Harvey Johnson, CO, an orthotist with Eno River Orthotics in Hillsborough, NC. “They could have made AFOs and used them on normal people.”

But Manee explained that some of the forces used in the research could not have been applied to a human.

“It would have hurt a person,” he said. “A cadaver doesn’t complain. We can use higher level forces to see things more clearly.”

For this preliminary study, using a real AFO with real straps was not feasible, Manee said.

“This is the first study looking at strap placement for control in an AFO,” he said. “Preliminary experiments like this have to eliminate as many variables as possible. This test is looking for the ideal location to apply force, so we made a force applicator that applied it at a single, clear location.”

Although therapeutic shoes also use dorsal straps, the new findings about dorsal strap placement on AFOs may not apply to footwear.

“They are very much different,” Dibello said. “The posterior height of the AFO creates an alignment that resists plantar flexion and which requires anterior strapping to apply the forces that create the resistance; a shoe does not.”

Manee acknowledged that a shoe can disperse the controlling force along the entire dorsum and, as such, is a different animal.

“We were looking at a single location scenario,” he said. “The fact that there are multiple force application points allows shoes to use less optimal locations but more of them, sort of a shotgun versus a sniper rifle—either one gets the job done.”

Comfort and compliance

Comfort is a key factor in patient compliance. If patients are in discomfort, for example, feeling pressure on underlying foot tendons, they may loosen their AFO straps, which makes the device less effective.

“If you try to position the patient into a range of motion that they cannot functionally achieve, they will have tendon irritation from the strap, as well as increased motion or ‘pistoning’ within the AFO,” Jinright said.

Dibello, however, theorized that if the length of the resistant lever is maximized and the force applied accurately, that should minimize pressure at that point.

Manee was less inclined to think that patients fuss with their straps.

“Yes and no,” he said. “If it is painful, a patient will without a doubt loosen the straps. That said, straps are not usually painful, and if they are, adjustments to the location are made. The only patients who sometimes pull their straps are young children, but that is arguably less because the strap hurts and more because they just want the brace off as it is annoying and the pressure from whatever correction it is providing can be uncomfortable.”

An area in which practitioners seemed to agree was that the size and strength of the patient doesn’t matter with regard to strap placement. The thickness of the plastic or the rigidity of the material should compensate and accommodate for the patient’s size and perceived strength.

“I think it would be logical that, as the patient’s height increased, I would expect his foot length to also increase, so to use the formula described here would optimize the placement of the strap regardless of the size of the patient,” Dibello said.

As to whether the strap placement would depend on the type of correction being made, Dibello said, “As long as your device is designed to reduce plantar flexion of the foot, then this formula for determining its placement should optimize your result.”

Future directions

There are a multitude of other issues surrounding strap placement, including how the ends of the strap are attached to the device, the type of material used (eg, plastic, carbon fiber), the design of the device (eg, solid, hinged). For example, the AFO design might dictate where the strap goes since the attachment point has to be in a place where it can be secure, not cause discomfort to the patient, and avoid ankle joints. Those factors were outside the scope of the current Georgia Tech study, but may be addressed in future ones.

“Now that we know the angle and we know that the anatomy is a key role in that measurement, which is the biggest finding that we had—or let’s say something that wasn’t as intuitive to us clinically before the study we’ve done—now I can start there and continue with this line of research,” Kogler said.

He pointed to another set of conditions—replacement of the steel bar, for example, or the influence of the surface area, foot geometry, or tendon tightness in front of the ankle—that will need to be examined.

“All of these variables come into play with us as clinicians and all of those will have to be addressed. But there will be additional studies that will be a follow-up to this,” Kogler said.

For example, he doesn’t believe current strap utilization, while cheap and functional, is optimal.

“I think there is a better system to limit the movement,” he said.

Kogler theorizes that an interface that more closely matches the anatomy of the individual user, perhaps a custom half shell, would be optimal. He also said research must address understanding how that force is applied through the anatomical structures, ie, tendons, in front of the ankle.

“That may change how that force is applied,” he said. “When those tendons are tight because the muscles in the middle of the leg are firing, those tendons are more like cables and they move away from the bone. That’s not very comfortable. They’re not meant to have pressure.”

Because of the limitations of this study, Kogler is eager to continue to investigate to find the best way to control foot movement inside an AFO. In their next study, Kogler and colleagues will look at the anatomy and make a device that contours over the dorsal aspect of the foot and ankle. The AFO would be the same as the devices used in the earlier studies, but the strap might have a custom-made pad or curved plastic.

“Working with how the anatomy functions to make something that fits better, I think is going to be more optimal,” he said. “We’re going to be taking MRIs of the foot in that area, identifying what happens to those structures when the muscles fire, and starting to look at a new design of a device that more effectively controls that movement.”


1. Manee T. Optimal dorsal strap placement and angulation to prevent pistoning in an ankle foot orthoses. Presented at the American Orthotic and Prosthetic Association National Assembly, Las Vegas, September 2011.

2. Kogler GF, Conne KM, Manee T. Optimal placement and angulation of a dorsal motion control mechanism to resist heel rise in an orthosis. Prosthet Orthot Int 2014 June 11. [Epub ahead of print]

Leave a Reply

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.