Entrepreneur . Researcher . Physician . Specialist

Home » Blog


bottom image 2 BLOG PAGE

Exercise is a cornerstone of treatment for diabetes, but for years patients with peripheral neuropathy have been dis­couraged from weight-bearing exercise for fear of further increasing ulceration risks. Now new research is turning that advice on its head.

By Cary Groner

A classic case of innovative research turning conventional wisdom on its head is changing the way clinicians approach exercise in patients with diabetic neuropathy.

For decades, patients with type 2 diabetes and peripheral neuropathy were cautioned against weight-bearing exercise out of fear that the accompanying stress on the foot would lead to plantar ulcers. Then, in 2003, scientists began to report surprising findings.

“Prior to those studies, the feeling was that weight-bearing exercise was too risky to recommend to patients who lacked sensation,” said Joseph LeMaster, MD, MPH. LeMaster, for many years an associate professor in the Department of Family and Community Medicine at the University of Missouri, will move to the University of Kansas this fall. “There was evidence that people with neuropathy had increased plantar pressures, and those were considered an independent risk factor for foot ulcers.”

In 2003, LeMaster and his colleagues published a study of 400 diabetes patients with a history of foot ulcers and found that increased weight-bearing activity didn’t increase the risk of reulceration. Moreover, the most active subjects saw the most significant risk reduction, and the effects were the same regardless of whether subjects retained foot sensation.1

That same year, researchers from Washington University in St. Louis reported in Clinical Biomechanics that diabetes patients with a history of plantar ulcers were 46% less active and accumulated 41% less daily stress on the forefoot than nondiabetic and diabetic control subjects without a history of such ulcers.2 At first, the finding seemed so counterintuitive that people weren’t sure what to make of it. The authors ultimately concluded, conservatively, that subjects with a history of plantar ulcers were susceptible to injury at relatively low levels of tissue stress.

These studies flung open the door to further investigations, however. In 2004, scientists confirmed in Diabetes Care that neuropathic patients who exercised more had lower rates of ulceration than those who were relatively sedentary.3 Two years after that, in 2006, researchers in Italy reported that, far from being deleterious, exercise could help prevent neuropathy’s onset or modify its natural history.4 Right on cue, then, in 2008, Washington University researchers reporting on the Feet First study noted that promoting weight-bearing activity did not lead to significant increases in foot ulcers.5 Finally, in 2010, the American Diabetes Association, together with the American College of Sports Medicine, acknowledged this accumulating body of evidence and published new guidelines that, for the first time, endorsed weight-bearing exercise for patients with diabetic neuropathy in the absence of foot ulcers.6

“The new guidelines represent a big change,” said Michael Mueller, PT, PhD, a professor of physical therapy at Washington University School of Medicine. “For the first time, people with diabetic neuropathy are explicitly encouraged to do weight-bearing exercise.”

Although this rhythmic chronology outlines what appears to be a straightforward investigation that changed medical practice, the story is more nuanced. A number of questions have bedeviled researchers, and continue to. For example, what’s the chicken and what’s the egg? That is, do people get more ulcers because they get less exercise, or do they exercise less because of their ulcer history? Or, for that matter, are other variables involved that no one yet understands? These and other issues, such as how to distinguish those at highest risk of ulceration from their peers and how to adjust exercise regimens accordingly for individual patients, are only now starting to become clear.


Back in 2002, Mueller published a paper in Physical Therapy whose relevance to this issue was not immediately clear, but which turned out to have a big impact. In that article, he proposed a “Physical Stress Theory” (PST) of tissue adaptation, the premise of which was that changes in the relative level of physical stress cause a predictable adaptive response in biological tissues.7 In a nutshell, the theory suggests that tissues respond to stress in predictable ways: stress levels that are too low lead to reduced stress tolerance and atrophy; mid-level stress produces no change; moderately high levels increase tolerance; and too much stress leads to injury and tissue death. The goal for practitioners seeking to increase their patients’ strength and resilience was to identify the levels that increased tolerance and work carefully from there.

Mueller also made several points that affected later researchers:

  1. Stress exposure is a composite value comprising magnitude, time, and direction of stress application.
  2. Extreme deviations from the maintenance stress range have serious consequences.
  3. Individual stresses combine in complex ways to contribute to the overall level of stress exposure, and tissues are affected by the history of recent stresses.
  4. Excessive stress can arise due to a brief, high-magnitude stress; a long duration of low-magnitude stress; or a repetitive application of moderate stress.
  5. Inflammation occurs immediately after injury, reduces the injured tissue’s stress tolerance, and requires that the tissue be protected from further stress until the inflammation subsides.

Many of these points turned out to be crucial to understanding how to manage diabetic neuropathy in the context of exercise.


The lead author of the 2004 study in Diabetes Care was David Armstrong, DPM, MD, PhD, professor of surgery and director of the Southern Arizona Limb Salvage Alliance (SALSA) at the University of Arizona College of Medicine in Tucson. An important aspect of his team’s findings was not just that more active subjects were less prone to ulcers, but that variability in activity was an important predictor of ulcer risk. Eight of 100 patients with diabetic neuropathy ulcerated during the average evaluation period of 37 weeks, and although they were significantly less active than those who remained ulcer-free, there was also much more variability in their exercise levels, as measured by high-capacity computerized accelerometer/pedometers.

“People who had wide swings in activity were at greater risk,” Armstrong told LER. “An example would be someone who’s not very active, then suddenly remembers their grandkid’s birthday and leaps off the couch, runs to the car, then spends an hour and a half walking around the mall. They do more in a couple of hours than they usually do in two days.”

When Armstrong and his colleagues first evaluated their data, they were flummoxed.

“We sat there wondering what was going on,” he said.

Their conclusion, however, echoed Mueller’s observations about the importance of tissue stress levels and the consequences of extreme deviation in them.

“We believe what we’re seeing is that it’s just like a lot of other places in the body,” Armstrong explained. “If you don’t use it, you lose it. If skin is allowed to atrophy, then maybe it’s weaker than skin that’s getting tenderized, as it were, by frequent activity.”

Armstrong noted that patients must be monitored carefully, as they were in his study, and that exercise has to be optimized for the individual.

“People can’t run a marathon with profound neuropathy, but we’d like to try to train them so they could slowly become more active,” he said. “We want to dose activity the way you’d titrate a drug.”

As for the chicken-and-egg problem—which comes first, the ulcer or the lower activity levels?—researchers are continuing to probe the reasons first ulcers appear. Manish Bharara, PhD, a research assistant professor at SALSA and a colleague of Armstrong’s, speculated that overall control of blood glucose levels may affect the resilience of damaged tissues.

“In diabetes patients, metabolic control affects other aspects of physiology, and could affect the quality of the tissue that is regenerated as someone heals,” he said.

A couple of Armstrong’s earlier papers may shed light on the issue, as well. In a 2001 article in the Journal of the American Podiatric Medical Association, Armstrong and his colleagues noted that diabetic patients with a history of neuropathy or ulceration took more steps per day inside the home than outside, and that only 15% of them wore their prescribed footwear inside.8 A paper in Diabetes Care  in 2003 reported that subjects with foot ulcers wore their off-loading devices for only a minority of steps taken each day.9 Noncompliance with preventive footwear or curative devices could conceivably be similar in effect to low activity levels, then, in that both are associated with ulceration and poor healing. One possible explanation is that, compared to high-activity patients, low-activity patients are taking significantly fewer steps per day in footwear designed to help their feet avoid injury or heal (activity studies have not consistently reported compliance data).

“It even turns out that sometimes just standing for long periods can be potentially dangerous,” Armstrong noted.10 “This is all about better identifying risk and helping us better coach activity. We’re trying to get people moving, and in a lot of ways, that’s how we measure success.”

Individual cases

The Feet First study made it clear that clinicians must carefully consider the patient’s history when prescribing exercise, according to lead author LeMaster.

“In that study, we felt that the exercise program, combined with the careful monitoring we conducted, showed that the benefits of exercise outweighed the risks,” he said. “But it’s quite another thing to say that people who have lots of recent foot ulcers should go out and do this. A good percentage of the people in the study had had prior ulcers, and we didn’t find that to be a predictor [of ulceration during the trial]. But we restricted people from walking if they had any breakdown during the study.”

People with a history of frequent and recurrent ulcers, he added, should be viewed in a different category than those included in the research. Furthermore, the study’s subjects had their feet examined weekly by a physical therapist for the first three months, and had a hotline to call if they showed signs of ulceration later.

Mike Mueller, a coauthor of the 2008 Feet First paper, likened the evolving view of exercise in those with neuropathy to a similar evolution in thinking about exercise in cardiac patients a few decades ago.

“There was a time when the prevailing opinion was that if you’d had a heart attack, you should not exert yourself,” Mueller said. “We came to learn that if you monitor the heart carefully and keep it within a safe range, exercise is beneficial. It’s similar with the neuropathic foot, although we’re still learning what the guidelines should be.”

Adjusting exercise programs to the individual based on variables such as ulcer history is still an emerging field, he noted, and based both on the evidence provided by research and on clinical experience.

“I believe that once you’ve had a full-thickness ulcer, you’re in a whole different category,” he said. “Even a mild one sends up a red flag that you’d better watch this person. There’s so much heterogeneity in the group of people who have diabetes and neuropathy that the program really needs to be tailored to the individual.”

Joint biomechanics

Part of the problem with such tailoring is that only recently has research begun to describe the relationship between biomechanics and diabetic neuropathy.

For example, a 2007 paper in the Journal of Applied Biomechanics found that diabetic subjects with neuropathy had stiffer ankles than diabetic subjects without neuropathy.11 It’s known that normal mobility allows the foot to flexibly dissipate impact, then become rigid during push-off.12  Restricted mobility in the foot and ankle joints, then, could hinder this transition and contribute to abnormal plantar loads.13

Citing such evidence, Smita Rao, PhD, an assistant professor of physical therapy at New York University, published a paper in 2006 outlining how changes in muscle could account for decreased range of motion (ROM) and increased stiffness in patients with diabetes.14 In a subsequent article in Gait & Posture, she and her colleagues reported that decreased sagittal motion of the first metatarsal and lateral forefoot and frontal motion of the calcaneus were key elements that could contribute to increased, sustained plantar loading in patients with diabetes and neuropathy.15

“There’s a big push to emphasize exercise in patients with diabetes and peripheral neuropathy, but those patients are also at higher risk for tissue breakdown, so I wanted to explore the mechanisms that put them at risk,” Rao told LER. “We showed in the Gait & Posture paper that a lot of these patients try to reduce the effects of their stiffness by walking slower and taking shorter steps. When I examine them, I want to look at ankle range of motion, all the mechanical factors that may affect tissue breakdown; but I also want to assess how they walk, find focal regions of high pressure, then put those two together to see if walking is the best activity for this person. Some might need protective footwear, and some should ride a stationary bike instead.”

In her current research, Rao and her colleagues at NYU are examining ways to bring a number of fields together.

“My grandfather had diabetes, so I have a personal connection to the field,” she said. “All these negative effects begin with high blood sugar, so we’re trying to combine medical, surgical, and rehabilitative interventions in patients with diabetes and neuropathy.”

Exercise and balance

Other research has looked at the importance of augmenting exercise with balance training, which has been shown to improve clinical balance measures in neuropathic patients.16 A study published in Diabetes Care in 2010 demonstrated, moreover, that six weeks of such training reduced the risk of falls in 16 older patients with type 2 diabetes and mild to moderate neuropathy.17 In that research, exercise sessions included a balance/posture component (lower-limb stretches and leg, abdominal, and lower-back exercises) and a resistance and strength-training component using machines. The regimen led to better reaction times and affected sensory, motor, and cognitive processes, leading to a significant decline in risk of falls.

Lead author Steven  Morrison, PhD, director of research in the School of Physical Therapy at Old Dominion University in Norfolk, VA, told LER that his group’s work was motivated partly by the fact that older diabetes patients’ risk of falling is 10 to 15 times that of healthy age-matched controls, which affects their confidence and ability to exercise.

“To be balanced, you need a certain amount of strength and a certain amount of coordination,” he said. “We found that after six weeks of training, type 2 diabetic individuals become more like the control group—there’s very little difference in terms of how much they sway and what their balance is like.”


David Sinacore, PT, PhD, a professor of physical therapy and medicine at Washington University, and one of the researchers involved in studies of exercise and neuropathy there, emphasized that monitoring—by clinicians or the patients themselves—is crucial to successful exercise programs in those with diabetic neuropathy, particularly if they also have foot deformities such as those resulting from Charcot arthropathy.

“I’m a firm believer that these folks need to exercise for their diabetes,” he said. “But if they start to develop lesions, they need to be addressed.”

Of course, as most clinicians know, there is often a gap between ideal and real-world monitoring levels.

“It’s hard for these patients to check the bottom of their feet regularly, so they sometimes don’t do it,” Sinacore said.

One way to help is with temperature monitoring. Sina­core recommends foot-temper­ature gauges that patients can use right after exercising, some of which are hook-shaped to ease plantar access.

“When we monitor them here, we check temperature before and after exercise,” he said. “We’re looking for hot spots and temperature dif­ferences that may indicate that they’re developing a lesion.”

In such cases, therapists recommend that patients de­crease their exercise levels for a while and have their footwear modified to relieve pressure.

David Armstrong agreed that thermometry provides a way of keeping track of the damage caused by weight-bearing exercise.

“We want our patients to dose their activity by checking their skin temperature just as they dose their insulin by checking their glucose,” he said.

His colleague, Manish Bharara, conducts innovative research in this aspect of care.18

“In the last decade we’ve learned that a four-degree difference between two similar sites on both feet is an ulcer risk,” he said. “If the pattern persists over multiple days, the patient should reduce activity and immediately see a doctor.”

Bharara and his colleagues are developing a thermometry scale to address some of the inconveniences typically associated with measuring foot temperature at several sites. Patients stand on it—it’s similar to a bathroom scale—while it measures foot temperature at 20 sites on each foot and records the data. The scale speaks to the patient—telling him, for example, that his right big toe temperature is 5° warmer than the left. Moreover, if the scale detects an abnormal pattern for more than two days, it can be programmed to send a message to the physician’s office and make an appointment.

“Something like this could really help manage patients’ diabetes better, because the biggest barrier is compliance,” Bharara said.

New research

Other researchers are examining variables that affect neuropathic patients’ exercise capabilities, as well. For example, at the Center for Lower Extremity Ambulatory Research at Rosalind Franklin University in Chicago, Bijan Najafi, PhD, associate professor of applied biomechanics, has studied factors including gait initiation in this context.19 As opposed to the measures of steady-state walking—such as rate or number of steps—typically used in exercise studies, a prolonged gait initiation phase (the period between upright posture and steady-state gait) may be associated with increased fall risk.

“During the initiation of the step, there’s an important acceleration phase, and it creates a lot of resistive force,” Najafi said. “We’ve found that neuropathy patients have longer gait initiation. This makes sense, because to reach steady-state gait, people have to gather somatosensory feedback to find the speed at which they can walk safely and minimize energy costs. Neuropathy patients have impaired somatosensory feedback, though. But we believe that if we can provide a good exercise to compensate, we may be able to improve the gait initiation phase.”

One way to help, Najafi thinks, is to take a cue from the dance world.

“If you’re trying to explain a movement problem to a dancer he may not get it, but if you put a mirror in front of him and show him the correct position of the joints, he can improve his motor skills,” he said. “The brain is plastic, and if it realizes there’s an error, it will try to minimize it next time. So we hope that by letting neuropathy patients observe their errors this way, they may improve their motor skills.”