oandp.com

Report from the Academy Meeting in Orlando: Part Two

Kim De Roy , a PT and P&O from Iceland who works for OSSUR showed several video segments of the Victhom powered prosthetic knee currently under development. Although it was tethered by a cable in most of the clips, it appeared to be totally self-contained in at least two of them. It reportedly has a similar mass and weight to a human leg, which would be 2-3 times heavier than current components in its current configuration. However, the 12 beta test subjects reportedly did not find the weight objectionable, perhaps because the motor actively flexes and extends the knee without added effort by the amputee. And, there is some scientific evidence that momentum from a shin section approximating the weight of the normal shank may help "pull" the person forward during swing phase more readily than a light weight shin.

Video observational gait analysis suggested that the overall motion when walking was smooth and relatively symmetric with the opposite limb. Stair and ramp climbing seemed well controlled and approximated normal speed. The overall subjective evaluation of this cohort of subjects was generally positive and encourages further refinement of this development.

De Roy suggested the term "active concentric torque" to denote such actively powered knees, and speculated that this feature might have particular value for weaker amputees. This may be a useful distinction since a motorized component [such as the more familiar electric hand] uses electrical energy to simulate concentric contraction of the missing muscles that formerly actively moved the body part. Currently available prosthetic knees with swing phase control would then be considered "passive concentric torque" devices, whether microprocessor controlled or not, since they simulate the eccentric contraction of missing muscles that would have decelerated the limb.

Margrit Meier, PhD , a Swiss prosthetist-orthotist researcher at Northwestern University, presented preliminary data from an ongoing study she is conducting to learn more about differences between the C-Leg and non-microprocessor controlled [N-MPC] knees with stance stability features. To date, Margrit has studied 8 transfemoral and 3 knee disarticulation amputees negotiating an obstacle course with and without a mental load task to distract them. The mental load consists of counting backwards out loud on command in different increments.

This is a cross-over study where each subject is fitted with three different knees in randomized order, with a four week acclimatization period prior to testing. The components being compared are:

• The monocentric Otto Bock C-Leg with microprocessor-controlled hydraulic stance and swing flexion and extension resistances
• The polycentric Otto Bock 3R60 knee with N-MPC stance flexion and hydraulic stance extension and swing resistances
• A monocentric knee frame with an OSSUR Mauch SnS hydraulic cylinder providing N-MPC stance flexion and swing phase resistances

Early trends were sometimes counter-intuitive, with the mental load condition having only a small effect and the 3R60 tests resulting in completion of the obstacle course 2% faster than without the mental load. Margrit speculated that the inherent stability of the 3R60 geometry might offer an advantage on the yielding surfaces of this obstacle course.

One possible confounding factor was that several subjects used the SnS in their daily prostheses, some for many years, so they may have been more adept at taking advantage of its performance than the new knees. However, Margit reported that she had done a sub-analysis of this cohort and did not detect any marked differences, so this concern may not be significant. Once the targeted 15 subjects have been enrolled and tested, Margrit will analyze and publish her results in more detail. This will be a very interesting paper due to the well-controlled nature of this study.

Scott Elliott CPO reviewed some of the subjective evaluations from beta testers who walked on the OSSUR Rheo MPC knee. OSSUR developed a visual analog scale to record the subjective opinions of the amputee subjects on a number of parameters. Their responses were generally favorable comparing the Rheo to their current N-MPC knee except in three areas: weight, kneeling, and the need to recharge daily. The video clips of amputees walking with the Rheo showed a smooth, symmetric, and confident stride and controlled foot-over-foot descent of stairs and ramps.

Doug Smith MD reported on some of the difficulties encountered in the Prosthetic Research Study investigation of the performance of the C-Leg, emphasizing how challenging it is to define valid objective measurements that are sensitive enough to detect the differences that patients report subjectively and clinicians believe are important. Their protocol is to recruit 20 experienced ambulators with amputations at the knee or above, with half being rated as K2 according to Medicare and half K3. Each subject received a clone of their current socket, with either a MPC C-Leg or N-MPC knee in random order. They were allowed 3-7 months to acclimate to each knee, and received gait training to optimize their function with both components. They then completed a battery of tests that included the PEQ, a timed obstacle course, and instrumented gait analysis.

In general, the PEQ showed higher patient satisfaction with the prosthesis having the MPC knee, and an increase in dissatisfaction with the N-MPC knee once they had experience with the C-Leg. Doug cautioned that it was very difficult to accurately measure stumbles, falls, and the mental load of using a prosthesis. He also noted that several patients that even after as many as 9 months of gait training and full time use of the C-Leg, they still felt that their ability to ambulate was continuing to improve. This raises the question of whether MPC knees take longer to master than N-MPC knees, perhaps due to the number of added biomechanical features they can offer compared to simpler components. This is an ongoing study that will be published in the future.

This session concluded with a practical review by Ryan Blanck CPO of how to define and explain the functional and medical necessity for MPC knees.

This free paper session was one of the best I have heard on this increasingly important topic, and all of the presentations were directly relevant to contemporary clinical practice. It was noteworthy that, with one exception, all of the speakers were from overseas. This reflects the lead our European and Scandinavian colleagues have in applying evidence based practices in P&O rehabilitation, which is facilitated by their centralized health care funding systems.

Elizabeth Condie PT started this session by reviewing some of the amputee statistics gathered in Scotland over the past ten years as a result of the efforts of a rehab research group who defined a series of outcomes measures that are now mandated by the Scottish government. The legal mandate for all the clinics to collect and report the same data means that Scotland can accurately determine trends in their amputee population. Overall prosthetic outcomes are measured by use of the Locomotor Capabilities Index. They no longer use preparatory devices but instead start with a custom prosthesis an average of 35-40 days post-amputation.

Over the past decade, the ratio of transtibial to transfemoral amputations has gradually decreased to approximately 1:1. This is a marked difference from earlier decades when the reported ratio was 2 or more TTs for every TF. The reason for this change is not known but will be the subject of future research. Furthermore, the Scottish data show that the rate of infection and of revision to a higher level is also increasing. Perhaps due to the generally older age and larger number of comorbidities of today's new amputees, only 65% of transtibial amputees and 25% of transfemoral amputees even receive a prosthesis.

Anton Johanson , a P&O from Sweden, reported somewhat different figures from the Scottish experience. The median age at amputation is now 81 years old, with a re-amputation rate of 12% and a one-year mortality rate of 27%. The ratio of TT to TF levels is 2.5:1, with 68% being ambulatory prior to amputation. Overall, 41% of all new amputees received a prosthesis. However, if those who died within 30 days of amputation are excluded, 70% of the remaining cohort were fitted.

Johanson has developed a detailed protocol for transtibial amputation that begins with a removable rigid dressing applied 5-7 days post-operatively, followed by use of a roll-on silicone liner for compression therapy until fitting. The initial prosthesis is always the direct-formed ICECAST device. At his hospital, all amputations are done only by orthopaedic surgeons using a sagittal flap method. He reports that their comprehensive approach to amputee rehabilitation has resulted in the lowest rate of amputation combined with the highest fitting rates in Scandinavia.

OSSUR will soon offer a commercial version of their rigid dressing method, which looked quite practical to me. The basic approach seems to be protecting the suture line with a clear occlusive dressing and rolling on a thin silicone liner. A bead-filled cylinder is then applied, gently hand molded, and evacuated until it becomes rigid. Releasing the vacuum allows easy removal of the dressing for wound inspection.

Johanson emphasized the value of early walking on a prosthesis, even if there is not complete healing. This is a philosophy that has always been successful in my hands, both in Chicago and particularly during my years at Duke University. Many experts believe that the intermittent compression from controlled walking in a well-fitting socket hastens healing, perhaps by promoting circulation, although I am not aware of any conclusive evidence supporting this practice. Mr. Johanson showed an impressive case series in which a gaping open area surrounding the distal tibia gradually closed in over time while the patient completed gait training and used the prosthesis daily.

Mark Geil PhD reported the results of a nice basic study investigating the accuracy of manual anthropomorphic measurements typically taken in the course of amputee fitting. He compared the results from students in an entry-level program to experienced practitioners from the community and found similar accuracy, although the students' results were slightly better. All the instruments he investigated, including cloth tape measures, length and diameter gauges, et cetera were accurate within 1/10 of a millimeter. The one exception was the VAPC caliper, which had a larger error rate, perhaps because it involves subjectively compressing the popliteal soft tissues "correctly". In the second phase of this study, Mark will look at non-contact measurements using CADCAM input devices.

Gudjon Karason introduced the Patient Activity Monitor [PAM] from OSSUR, which is a step-counting device similar to the more familiar Step Activity Monitor developed at PRS. He pointed out that these validated devices can record number of steps taken and calculate velocity and similar aspects of gait over an extended period of time. As the cost and complexity of using these instruments decreases, it is likely that they will be used more often in the clinical setting.

Magnus Lilja followed up by presenting his initial research with the PAM, noting that commercial pedometers have been shown to underestimate the actual number of steps taken by as much as 75%. The PAM is programmed to analyze amputee walking into a "PAM Index" that estimates the ratio of time spent sitting, in limited walking, and sustained walking activities. His preliminary data suggest that amputees walk much less often and less distance than non-amputees, even though they may consider themselves to be "highly active".

Catarina Lowenadler concluded the outcomes presentation by reviewing a comprehensive amputee management program developed in Scandinavia with the acronym of "S.M.A.R.T.", which seems very well organized and integrated. The price data she has gathered in Sweden suggests that the cost for the artificial limb amounts to 6% of the total rehabilitation expenditure per patient.

The first free paper in this group was a very nice practical review by Dennis Janisse CPed of some of the techniques he uses to successfully manage the severely deformed Charcot foot. In most cases, Dennis has not found it necessary to provide custom-made shoes. Instead, he prefers to substantially modify extra depth shoes to accommodate fixed deformities. One useful technique he calls "drill and fill", which involves using a hole saw to remove material under the lesion, even if located deep inside the toe box, by drilling directly through from the sole. The resulting excavation is filled with varying densities of polyurethane RTV [PQ] and the sole plug glued then back in place. Dennis also showed his method to widen a commercial shoe as much as ¾ inch by slicing, spreading, and re-gluing the sole. The result accommodated marked foot deformity but looked more normal than the typical custom molded shoe and was far less expensive to produce.

Pierre Leung MSME reviewed recent modifications to the Rancho Mobile Arm Support that have made this useful aid for people with proximal upper limb weakness from quadriplegia much sleeker in appearance and easier to adjust for the clinician.

Jessica McFarland CO presented her ambitious resident research project on ethical dilemmas in clinical practice that involved surveying a random sample of 500 certified orthotists and 90 residents. This was a very nicely done student project that I hope will be written up in detail and submitted for publication.

Roger Weber CO and Ronan Reynolds demonstrated their concept for a "vertebral orthosis", which is an AFO, KO, or KAFO having linked segments that mimic the controlled motion of the human spine. The AFO has a posterior strut that is carbon fiber whose resilience can be altered to change the resistance to plantarflexion of the foot segment. The anterior elements are linked with elastic members that return the foot to neutral in swing phase. During rollover, as the "vertebral" elements close anteriorly, the strut effectively shortens and the resistance to motion gradually increases.

Varying the spaces between the elements, as well as geometry and material for the posterior strut or the anterior members, allows the orthotist to change the overall resistance profiles. Their force-deflection data demonstrated that they could design an AFO with an ankle resistance curve that very closely mimicked that of the normal ankle-foot complex. The KAFO application could also simulate knee flexion in early stance, and seemed well suited to combine with a stance control orthotic knee joint. This is a very intriguing approach that offers a novel method to "fine tune" range of motion or resistance of custom-made orthoses.

Steve Irby MS provided an update on the stance control orthotic knee joint being developed at the Mayo clinic that they call the "dynamic knee brace system". It uses various sensors to trigger the SCO function and appears to be able to disengage the stance resistance even when the knee is significantly flexed. This unique capability may mean that the Mayo development, which is now being commercialized, will work for patients whose knee flexion contractures preclude use of other SCOs. The data from 21 patients who have tried the Mayo device that Steve presented is the largest series of stance control orthosis results reported to date. Like the data from prior studies, Steve's results strongly supported the value of orthoses that stabilize the knee without disrupting swing phase.

Jack Uelendahl CPO started off this session by summarizing his extensive clinical experience fitting bilateral lower limb patients. His chapter in the recently updated Atlas elaborates in more detail on his views. One important observation was that, due to the continuing increase in amputations due to dysvascularity, the incidence of bilateral loss is also increasing.

In general, Jack's comments reflected the broad consensus that has emerged regarding basic principles for managing this challenging population such as: start with low profile preparatory limbs [including "stubbies"] and progressively increase the complexity of the devices as the patient masters walking on nearly full-length prostheses. He noted that the success rate with bilateral prostheses is generally favorable when both knees are preserved, and that most people with bilateral traumatic or congenital limb deficiencies do well even with higher levels of loss. Successful use of a prosthesis as a unilateral amputee has been shown to correlate with later success as a bilateral, so early rehabilitation in cases when contralateral amputation is anticipated may be helpful.

Jack also highlighted some areas of clinical disagreement, such as whether or not highly mobile ankle mechanisms were helpful for individuals with bilateral loss. His recommendation was to recommend stiffer ankle-foot components initially to foster standing balance and a sense of stability, transitioning to more flexible choices that may make walking easier once the patient's balance has improved. Jack is a strong proponent for shock absorbing and energy storing components for this patient group, but suggested that since shock pylons become shorter under weight bearing, it may be best to combine them with polycentric knees offering additional swing phase clearance.

Brian Ruhe PhD , from Northwestern University, analyzed gait analysis done with five bilateral transfemoral amputees walking on prostheses. All were long-term community ambulators with artificial limbs, and each one used at least one balance aid when out of the home. Observational video gait analysis and instrumented laboratory data both confirmed the widely varying gait patterns among this group of patients. Many of the differences were directly related to the choice of components in the prostheses. Brian's preliminary work supported the clinical impression that both prosthesis design and rehabilitation outcomes for these patients are highly individualized.

Steve Gard PhD , also from Northwestern, reported on his research into the effect of multiaxial ankles and transverse plane torque absorbers on the gait pattern of bilateral transtibial amputees. This well constructed crossover design began with use of a dynamic response foot [Seattle Lightfoot II] for two weeks and the collection of baseline data. On a randomized basis, subjects then received either a multixial ankle [Endolite Multiflex] or a torque absorber for two weeks. The components were again changed so that all subjects had used all component configurations for two more weeks. In the final stage of this investigation, all subjects had prostheses with all three components: DR foot, MA ankle, and torque absorber.

For the transtibial group, the temporal and spatial parameters of gait did not change significantly although subjects generally preferred the components offering the greater range of motion. Hip hiking was a common gait deviation for both those who lost their limbs due to vascular problems as well as those with traumatic loss.

Steve then repeated the protocol with 4 transfemoral subjects. Their temporal and spatial data was also relatively unchanged, except that the impact of the torque absorbers was more pronounced. Three of four preferred the prosthesis with all of the components, reporting greater comfort and ability to walk longer with the multiaxial dynamic response ankle-foot combined with the vertical shock absorbing pylon.