A Closer Look At Gait Analysis In Patients With Diabetes

Michael DeBrule, DPM

Observing the gait of patients with diabetes can reveal valuable information that can help avert complications. This author discusses gait abnormalities in patients with diabetes, reviews the influence of ground reactive forces and provides a practical guide to analyzing gait in this patient population.

If one considers observational or advanced gait analysis for podiatry patients with diabetes mellitus, there is no consensus for an approach and evidence-based guidelines are lacking. Therefore, it would not surprise me if a podiatrist casually glanced at a diabetic gentleman’s cane in the corner of an exam room but did not bother to watch him walk. This is a sadly missed opportunity to obtain important information that might help prevent limb loss.

   Let us take a closer look at gait analysis in patients with diabetes by attempting to answer the following questions:

• How do patients with diabetes walk?
• What factors alter the gait of people with diabetes?
• What about ground reactive forces?
• How should physicians perform diabetic gait analysis?

How Do Patients With Diabetes Walk?

Gait analysis studies for patients with diabetes vary on their inclusion criteria, the definition of neuropathy, walking surface, data collection methods and sensor technologies. Results from these studies can be confusing or contradictory. However, two clear trends emerge: decreased walking speed (velocity) and an increased base of gait (step width measured from one heel to the other, perpendicular to the line of progression).1-4 If your patient with diabetes walks with both decreased speed and an increased base of gait, these gait characteristics should serve as red flag warnings for advancing neuropathy and foot ulcer risk.

   Increasing the base of gait may increase stability and balance during ambulation.1 A wide base of gait is also present in children beginning to walk and along with some older adult gait patterns such as caution, cerebellar ataxia, choreic disorders (found in Huntington’s disease) and waddling.5 I have noticed in my patients that an increased base of gait is sometimes accompanied by cautious gait changes like abducted legs and arms, very slow walking, or careful turning. When this pattern occurs, I question the patient about anxiety, fall history, fear of falling and visual impairment.

   Researchers have also reported additional gait parameters in patients with diabetes.1,3,4,6-11 These include decreases in cadence, step length, single limb support time, the maximum vertical component of ground reactive force, plantarflexion moments, step variability, and knee and ankle mobility. Also, there are increases in double limb support time and time to complete the gait cycle. Furthermore, patients with diabetes may walk even slower on irregular surfaces like cobblestones.7 However, one should keep in mind that most of these additional parameters could be secondary to slower walking speeds. Slower walking speeds are associated with decreased ground reactive forces, decreased joint angles, decreased single limb support and increased double limb support time.12

   A recent study by Raspovic reported that patients with an ulcer history had significant sagittal plane decreases in range of motion for the ankles and first metatarsophalangeal joints.10 Otherwise, most gait analysis studies I reviewed had surprisingly little to say regarding sagittal plane abnormalities in patients with diabetes. However, I have often observed sagittal joint motion reversal associated with functional or structural hallux limitus in my own patients with diabetes. This pattern of gait changes, as documented by Dananberg, includes decreased great toe dorsiflexion during propulsion, delay in ankle plantarflexion (heel lift), delay or failure to achieve knee extension, decreased hip extension, straightening of the lumbar spine, and cervical flexion.13

What Factors Alter Diabetic Gait?

Although research has not firmly established causal relationships for gait changes in patients with diabetes, peripheral neuropathy remains suspect number one. Kanji and colleagues have suggested that large fiber neuropathy, which affects gait with the loss of pain sensation and proprioception, occurs often in the insensate foot and many patients are unaware of this condition.14 Accordingly, it is likely that a proprioceptive deficit causes patients with diabetes to walk more carefully than those without diabetes and at a slower speed. Study outcomes comparing patients with and without diabetes are not clear regarding which gait alterations are specific to peripheral neuropathy and which also affect those without neuropathy. However, it is most likely that gait abnormalities occur across the spectrum of diabetes and increase with disease severity.15

   There are some other interesting factors to consider besides neuropathy and proprioception. Research has suggested that decreased plantarflexor strength and ankle mobility contribute to a slower walking speed.9 Also, patients with a history of diabetes-related ulceration have less knee and ankle strength.10 Additionally, tissue changes can influence gait, foot-ground interface and the risk for ulceration in patients with diabetes.11 These tissue changes include poor skin quality, thickening of tendons and the plantar fascia, decreased joint mobility, muscle stiffness, and fat pad atrophy.

   While obesity is another consideration, most studies I reviewed did not comment on body mass index even though the patients in control groups usually weighed much less. It is well documented that increasing excess body weight decreases walking speed and increases double limb support.16 More studies are needed that compare patients with diabetes to those without diabetes with appropriate controls for gender, age and body mass index.

What About Ground Reactive Forces?

Ground reactive forces act in all three dimensions when we walk. The two major components of ground reactive forces include normal (vertical) forces associated with plantar pressure and shear (horizontal) forces. Most diabetic foot research has focused on the relationship of plantar pressure and ulcers, likely because commercial platforms, pressure mats and in-shoe systems are readily available. However, the measurement of shear force remains more mysterious and elusive. Shear is more difficult to measure and commercial systems are not available for office use.

   Plantar shear is an exciting new area of research and I think shear could one day prove to be more a significant factor for ulcer risk than plantar pressure. However, future studies are needed to substantiate a decrease in diabetic ulcers from shear-reducing shoe insoles or orthotic modifications. Another conundrum with shear is that we cannot eliminate it entirely. We need shear forces and ground resistance for normal walking to help propel us forward. If we eliminate too much shear, it might be similar to walking on one of the frozen lakes we have up here in Minnesota in January (helmet recommended).

   Plantar shear can cause calluses and recent studies have implicated it in the development of foot ulcers. Yavuz and coworkers found plantar shear values in patients with diabetes were significantly higher than in healthy controls.17 Also, for the majority of patients with diabetes, the sites of peak shear did not match the sites of peak plantar pressure. Lott and colleagues also reported increased maximal shear stress for patients with diabetic neuropathy and a history of ulceration.18

   Elevated plantar pressure is a moderate risk factor for ulceration in patients with diabetic neuropathy yet abnormal foot pressures alone do not seem to cause ulceration.11 A study by Masson and colleagues comparing the frequency of elevated plantar pressures in patients with diabetes and patients with rheumatoid arthritis found a similar frequency of elevated plantar pressures for both groups.19 About one-third of the patients in the diabetes group had a history of ulceration in comparison with none in the rheumatoid group. Furthermore, I have personally observed that patients with normal plantar pressures can still get an ulcer while patients with elevated plantar pressures may not ulcerate.

   Attempts to identify a peak pressure threshold to predict diabetic ulcers based solely on plantar pressure have been disappointing. There is no agreed threshold value at this time.20 Armstrong and colleagues did not find an optimal cutoff for peak pressure in patients with diabetic ulceration and another study by Lavery and colleagues observed a sensitivity of 64 percent and specificity of 46 percent using an 87.5 N/cm2 optimal cutoff value.21,22 Furthermore, Veves and coworkers reported that only 38 percent of peak pressure locations matched the area of ulceration for their patients with diabetes.23 They also observed that the location of peak pressure changed in 59 percent of patients over 30 months’ follow-up.

   Historically speaking, we tend to view any treatment that lowers peak plantar pressures as good and increased plantar pressures as bad. For example, we commonly accept that diabetic shoe insoles can lower peak pressure by distributing pressure over an increased area. However, peak pressures might not be that cut and dry. What about patients with diabetes who increase their walking speed after a treatment intervention like an ankle foot orthotic (AFO) or Charcot foot reconstruction? As their speed increases, forces on the foot increase and so should peak plantar pressures. Is peak pressure as detrimental as we once thought or should we be looking at other gait parameters?

   Although other gait parameters like pressure-time integrals and pressure gradients have also received study in patients with diabetes, perhaps we should also consider parameters independent of walking speed. Najafi and coworkers explored the use of an alternative walking parameter independent of gait speed called the regression factor.20 The regression factor tries to represent the similarity of the actual pressure distribution with a normal distribution using a timescale normalization scheme. Regression factor values range from -1.0 to +1.0 and as the value increases positively, the patient is walking more normally. In their study of post-foot reconstruction Charcot patients, Najafi and coworkers found regression factors increased postoperatively, suggesting a transition to more normal plantar pressure distribution. Interestingly, the regression factor did not change significantly when the study participants increased their average speed by 29 percent but peak plantar pressure increased by 8 percent.

   Instead of placing so much emphasis on one gait parameter like peak plantar pressures, shouldn’t we try to look at the big picture for multiple parameters? Pressure mats and in-shoe pressure systems may provide valuable information like prolonged lateral forefoot loading, early hallux plantarflexion, timing of heel lift, force versus time curves, force-time integrals, symmetry analysis, center of force trajectories, etc. Considering all of these gait parameters may help identify abnormal gait patterns and assist with treating our patients. Unfortunately, there is no evidence-based best algorithm to follow.

Pertinent Tips On Performing Gait Analysis For Patients With Diabetes

If you do not have a room dedicated to gait analysis at your clinic or hospital, you may have to do the best job you can despite limited space. Treadmills do not take up much space and would be great to use for a marathon runner with diabetes. However, treadmills might be a fearful experience for patients who have severe peripheral neuropathy or use assistive devices like canes or walkers. Treadmills also decrease stride length and increase cadence.24

   Therefore, I prefer to observe patients with diabetic neuropathy free walking down a hallway (or a walkway if you are lucky enough to have one). Personally, I prefer hallways over cramped exam rooms. Consider going out into the hallway to observe your patient walk from the front and back. Then observe the patient from the side (sagittal plane) by standing just inside the exam room door while your patient walks back and forth in the hallway in front of you.

   Your goals for gait analysis will vary from patient to patient. For example, let us say you are thinking about applying a fiberglass total contact cast to offload an ulcer. You might have your patient simply try on a controlled ankle motion (CAM) walker, watch the patient walk and quickly assess walking speed and stability. However, you might choose to spend a lot more time analyzing that same patient’s gait if you were planning surgery or thinking about an AFO prescription. Here are some suggested goals.

• Identify gait dysfunction (limited hip flexion).
• Identify gait asymmetry (lateral trunk lean on left side only).
• Identify timing problems (delayed heel lift).
• Attempt to distinguish between primary pathology and secondary gait compensations (knee osteoarthritis and decreased knee flexion).
• Appreciate new diagnoses that are not evident on history, exam or X-rays.
• Reassess gait to determine effectiveness of interventions.

   Most podiatrists do not have access to a full gait laboratory that has a force platform with video vectors, an electromyography amplifier or a 3D motion analysis system. There are also some expensive new wearable sensors that are beyond the scope of this discussion.

   So what can you do with limited technology and time? I suggest a stepwise approach for gait investigation in order of increasing sophistication depending on what is available to you and how much information you need. Keep in mind that advanced equipment will require more time, money and support staff to maintain and support it.

1. Perform a quick observational gait analysis, simply noting anything that strikes you as abnormal including the use of assistive devices.

2. Ask your patient to change into shorts and a T-shirt to help visualize arm, leg and knee motion.

3. Perform observational gait analysis in a systematic fashion using a checklist.25-27

4. Use a stopwatch with a measured walking distance to record basic temporal-spatial gait parameters like walking speed, cadence and stride length.

5. Use simple video with a camcorder, handheld device or cell phone.

6. Perform video analysis with multiple camera views, gait analysis specific software and key angle measurements.

7. Use a plantar pressure mat system or force platform.

8. Use a plantar pressure in-shoe recording system.

   Once you have performed gait analysis, next comes the hard part. You have to come up with a treatment plan by considering all the significant deviations you observed and their most likely causes (some will not be podiatric). You may then choose to perform additional assessments like checking for knee or hip range of motion, ordering magnetic resonance imaging (MRI), performing a timed get up and go test for fall risk, or whatever you feel is indicated.

   Without sounding too preachy, please do not underestimate the contribution of limb length difference to asymmetry and remember that a custom foot orthotic will not cure everything. Consider discussing the gait changes you observed in your patient with other specialists and get their opinions. Orthopedists, physical therapists, endocrinologists, rheumatologists, vascular specialists and primary care doctors may all help improve the gait in your patient with diabetes and prevent limb loss.

In Conclusion

Patients with diabetic peripheral neuropathy tend to walk slower with an increased base of gait. Gait analysis may improve your diagnosis and provide additional information for preventing foot ulcers, limb loss and morbidity. Gait analysis may also help when planning conservative or surgical treatments for diabetic foot offloading, or identifying an unsteady patient at risk of falling.

   Dr. DeBrule is in private practice with Midwest Podiatry Centers in Richfield, Minnesota. He is board certified in wound care.


1. Brach JS, Talkowski JB, Strotmeyer ES, Newman AB. Diabetes mellitus and gait dysfunction: possible explanatory factors. Phys Ther. 2008;88(11):1365-74.
2. Goldman S, Poonai D, Kamal O, K Khan. Emphasizing proactive gait assessment in patients with diabetes. Pod Today. 2011;24(4):20-25.
3. Petrosky J, Lee S, Bweir S. Gait characteristics in people with type 2 diabetes mellitus. Eur J Appl Physiol. 2005;93(5-6):640-47.
4. Katoulis E, Edbon-Parry M, Hakan Lanshammar, Vileikyte L, Kulkarni J, Boulton AJ. Gait abnormalities in diabetic neuropathy. Diab Care 1997;20(12):1904-07.
5. Salzman B. Gait and balance disorders in older adults. Am Fam Phys 2010;82(1):61-68.
6. Allet L, Armand S, de Bie RA, Golay A, Monnin D, Aminian K, de Bruin ED. Reliability of diabetic patients’ gait parameters in a challenging environment. Gait Post. 2008;28(4):680-686.
7. Allet L, Armand S, de Bie RA, Pataky Z, Aminian K, Herrman FR, de Bruin ED. Gait alterations of diabetic patients while walking on different surfaces. Gait Post. 2009:29(3):488-493.
8. Courtemanche R, Teasdale N, Boucher P, Fleury M, Lajoie Y, Bard C. Gait problems in diabetic neuropathic patients. Arch Phys Med Rehabil. 1996;77(9): 849-855.
9. Mueller MJ, Minor SD, Sahrmann SA, Schaaf JA, Strube MJ. Differences in the gait characteristics of patients with diabetes and peripheral neuropathy compared with age-matched controls. Phys Ther. 1994;74:299-313.
10. Raspovic A. Gait characteristics of people with diabetes-related peripheral neuropathy, with and without a history of ulceration. Gait Posture. 2013; 38(4):723-8.
11. Wrobel JS, Najafi B. Diabetic foot biomechanics and gait dysfunction. Jnl Diab Sci Tech. 2010;4(4)833-845.
12. Andriacchi TP, Ogle JA, Galante JO. Walking speed as basis for normal and abnormal gait measurements. J Biomech. 1977;10(4):261-268.
13. Dananberg HJ. Gait style as an etiology to chronic postural pain. Part II. Postural compensatory process. J Am Podiatr Med Assoc. 1993, 83(11):615-624.
14. Kanji JN, Anglin RE, Hunt DL, Panju A. Does this patient have large-fiber peripheral neuropathy? JAMA. 2010;303(15):1526-32.
15. Khan T, Guberman R. Gait alterations associated with diabetic neuropathy. Low Extrem Rev. 2012;4(8) 47-52.
16. Blaszczyk J, Plewa M, Cieslinska-Swider J, Bacik B, Zahorska-Markiewicz B, and Markiewicz A. Impact of excess body weight on walking at the preferred speed. Acta Neurobiol Expo. 2011;71(4)528-540.
17. Yavuz M, Tajaddini A, Botek G, Davis BL. Temporal characteristics of plantar shear distribution: relevance to diabetic patients. J Biomech. 2008;41(3):556-559.
18. Lott DJ, Zou D, Mueller M. Pressure gradient and subsurface shear stress on the neuropathic forefoot. Clin Biomech. 2008;23(3):342-348.
19. Masson EA, Hay EM, Stockley I, Veves A, Betts RP, Boulton AJ. Abnormal foot pressures alone may not cause ulceration. Diabet Med. 1989; 6(5):426-428.
20. Najafi B, Crews RT, Armstrong DG, Rogers LC, Aminian K, Wrobel J. Can we predict outcome of surgical reconstruction of Charcot neuroarthropathy by dynamic plantar pressure measurement?-a proof of concept study. Gait Posture 2010;31(1)87-92.
21. Armstrong DG, Peters EJ, Athanasiou KA, Lavery LA. Is there a critical level of plantar foot pressure to identify patients at risk for neuropathic foot ulceration? J Foot Ank Surg. 1998;37(4):303-307.
22. Lavery LA, Armstrong DG, Wunderlich RP, Tredwell J, Boulton AJ. Predictive value of foot pressure assessment as part of a population-based diabetes management program. Diab Care. 2003;26(4)1069-1073.
23. Veves A, Murray HJ, Young MJ, Boulton AJ. The risk of foot ulceration in diabetic patients with a high foot pressure: a prospective study. Diabetologia. 1992;35(7)660-663.
24. White S, Yack HJ, Tucker CA, Lin HY. Comparison of vertical ground reactive forces during overground and treadmill walking. Med Sci in Sports and Exer. 1998;30(10):1537-1542.
25. Kirtley C. Clinical Gait Analysis: Theory and practice. Churchill Livingston/Elsevier. Philadelphia, PA, 2006.
26. Root ML, Orien WP, Weed JH. Normal and Abnormal Function of the Foot, Volume 2. Clinical Biomechanics Corp., Los Angeles, CA, 1977.
27. The pathokinesiology service and the physical therapy department. Ranchos Los Amigos National Rehabilitation Center. Observational Gait Analysis. Los Amigos Research and Education Institute, Downey CA 90242.

   For further reading, see “Understanding The Impact Of Gait Analysis” in the April 2003 issue of Podiatry Today, “A Closer Look At Case Studies In Gait Analysis” in the August 2005 issue or “How To Evaluate For Leg Length Discrepancy” in the June 2004 issue.

   To enhance your online experience, check out Podiatry Today on your iPad or Android tablet.

Add new comment