July 2018

The rise of the crossover: A chronicle of cars and prosthetic feet

iStockphoto.com 825075568

The so-called crossover foot is emerging as an attractive option in lower-extremity prosthetics, just as crossovers found their way into the automotive market: because they offer utility.

By Phil Stevens, MEd, CPO, FAAOP

Let’s begin with an unlikely but enlightening departure from the kind of practical, clinical topic to which readers of LER: Lower Extremity Review are accustomed: a look at a trend in automobile design and sales in the United States.

“With crossovers so popular, will the sedan go the way of the station wagon, marginalized into a niche appreciated by some but bought by few?” wrote automotive correspondent Rick Popely in a 2017 Chicago Tribune article. 1 Popely answered his own question: “Not quite, but there’s no questioning that crossovers are here to stay.”

This recent trend in US car sales is notable. Sales of Honda’s Accord and Civic models have been eclipsed by the brand’s increasingly popular CRV, for example. It was Toyota’s RAV4, not the Camry, that was declared the company’s most popular model in 2017. For the first time, the Nissan Rogue outsold the midsize Altima. Even Porsche, long known for producing high-end sports cars, realized nearly two thirds of its sales in crossovers.1

Popely concluded: “Crossovers drove new vehicle sales to a record 17.5 million last year and in the process bumped aside the midsize sedan as the family vehicle of choice.”1

According to Stephanie Brinley, senior analyst for IHS Markit Automotive, “what’s really driving [the trend] is that people find that utility vehicles fit well in their lifestyles.”1The reason for the shift, then, is appreciation of utility. And there, with a single word, is where this discussion drifts into the evolving market for prosthetic feet.

A recent cross-over clinical trial identified, first, constructs in which crossover feet might prove beneficial and, second, considerations that might identify patients who can benefit from the new technology.

The predicament of choosing a prosthetic foot

In the field of prosthetic rehabilitation, increasingly sophisticated prosthetic feet have been developed to expand the range of user activities: from walking to running to participation in competitive sports. However, as Morgan and colleagues2 recently observed, “each prosthetic foot design is generally optimized for performance across a narrow range of activities and may inhibit use in other areas.”

The most frequently encountered example of this phenomenon is limitations in the otherwise elegantly designed energy-storing feet (ESF) and running-specific feet (RSF). Through iterative designs, ESF facilitate an increasingly smooth transition through gait phases across a range of walking speeds but fail to provide optimal support and energy return during sustained running. In contrast, RSF leverage long, stiff carbon-fiber springs to enable sustained running and high-speed sprinting but, lacking the prosthetic heel mechanism and split keel common to many ESF, are ill-suited for day-to-day walking and standing. Especially when an RSF has maximized the length of the carbon spring through a posterior mount of the foot to the socket, the limitations of the foot translate to limitations of the entire prosthesis.

Morgan observed that “… active users often require a primary prosthesis with an ESF and a sports prosthesis with an RSF or other specialized foot to participate optimally in a broad range of activities.”2

The result is a scenario of inconvenience. Although combined use of “primary and sports prostheses may allow people to engage in a variety of low-, moderate- and high-impact activities… the cost, maintenance, and burden of alternating prostheses may restrict participation in occasional or regular physical activity for many prosthesis users.” Like drivers who seek out a vehicle that meets the range of utility needs associated with their lifestyle, “users have expressed a desire for feet designed to accommodate a wider range of activities,” Morgan concluded.

Enter the crossover foot

Accordingly, the field of lower-extremity prosthetics has witnessed a shift toward crossover-type foot (XF) models designed to broaden the range of activities a user can perform with a single prosthesis. Much like the shared conveniences of a sports–utility vehicle interior that is built onto a car, rather than a truck, platform, XF models combine key attributes of ESF and RSF.

Thus, XF combine the heel spring and split keel of most ESF with the extended carbon spring of the RSF (see Figure 1). The resulting foot type is designed to span a user’s possible needs: from standing and slow walking, through quick walking and jogging, to sprinting.2 Because XF is a newly developed foot type, however, evidence to support its use is limited. This article reviews recent publications that have examined the impact of XF on key clinical considerations.2,3

Randomized study of crossover feet

Figure 1. Crossover foot. This prosthetic foot type combines the heel spring and split keel of most energy-storing feet and the extended carbon spring of running-specific feet.

Under the direction of researchers at the University of Washington, a convenience sample of study participants was recruited from area practices in which clinicians were experienced in the fitting of XF. The 27 patient–participants who completed the trial were all ≥1 year removed from a unilateral transtibial amputation of non-dysvascular origin and were using a well-fitting prosthesis with either an ESF or XF. Mean age of participants was 42 years; they were just under an average of 12 years post-amputation, reported an average daily use of a prosthetic of 15 hours, and had, on average, a Socket Comfort Score of approximately 8.5 on a scale of zero (most uncomfortable socket imaginable) to 10 (most comfortable socket imaginable).2

Subjects were randomly assigned to 1 of 2 prosthetic feet: an Össur Cheetah® Xplore (an XF) or an Össur Vari-Flex® (an ESR). These feet were worn for 1 month of accommodation before an outcomes assessment. Each subject was then crossed over to the other prosthetic foot, which they accommodated to for another month before repeating the outcomes assessment protocol.2

Given that the XF were laminated directly to the posterior aspect of their socket, each condition required a custom prosthesis with duplicated sockets and a shared custom flexible inner socket. Interface liners and suspension were maintained through both study conditions.2

In an effort to assess the comprehensive experience associated with the 2 foot designs, a range of laboratory and community performance measures, as well as several self-reported outcome measures, were collected.

User-reported outcomes

In the automotive industry, such considerations as cost and fuel economy affect decision-making; however, cumulative sales largely represent the simple metric of consumer preference. When the economy is good and fuel prices are low, consumers purchase the vehicle they prefer.

Similarly, although discussion below will describe a range of validated self-reported measures and laboratory outcomes in this study of an XF, user preference provides a good overview of head-to-head comparative experiences associated with the 2 conditions. Here, the XF was found to be the preferred condition. Of 19 participants who were invited to declare a preferred foot type, 17 said they preferred the XF and 2 asserted no preference.2 Specific activities for which subjects reported this preference provide additional insight into the potential benefits associated with this foot type. Walking quickly was the most frequently identified activity, followed by running, walking on an incline, ascending stairs, and carrying a heavy load.2

PLUS-Mfindings. In addition to the question of preference, a series of self-reported measures identified other areas of benefit associated with the XF. The Prosthesis Limb Users Survey of Mobility (PLUS-M) is an increasingly popular survey instrument that invites users to report on the degree of difficulty they associate with a series of mobility tasks. Because PLUS-Mhas been indexed against scores from more than 1000 prosthesis users, raw scores can be converted to T-scores, which can, in turn, be converted to an individual’s relative self-reported mobility within the larger amputee community. In this study, the average PLUS-M T-score with the XF was 64.2—roughly 5 points higher than the score associated with ESF (59.3).2 Converting the T-score to the more familiar construct of percentile, these higher-functioning participants scored in the 93rd and 83rd percentiles among all lower-extremity amputees with XF and ESF foot designs, respectively.

ABC Scale. The related construct of balance confidence was assessed using the familiar Activities-specific Balance Confidence (ABC) Scale. In this 16-item scale, users are asked to rate their confidence that they will not fall or become unstable across a range of increasingly difficult tasks. Originally assessed on a scale of 1 to 100, a revised ABC has been simplified to a 5-point response scale (0 to 4) with higher scores indicating increased confidence. In this trial, mean ABC scores were significantly higher (p = 0.005) with the XF than with the ESF (3.5 and 3.2, respectively).2 Perhaps of less concern in this higher-functioning population, the construct of balance confidence is significant because it has been found to be more closely associated with motor capability, performance, and social activity than related constructs of fall history or fear of falling.4

PROMIS-F. Fatigue was assessed using an instrument from the Patient-Reported Outcomes Measurement Information System (PROMIS), a library of robustly developed self-reporting instruments that include a questionnaire specific to fatigue (PROMIS-F). Consistent with other PROMIS measures, the T-score of the PROMIS-F is normalized such that a score of 50 represents the mean of the general population of the United States with respect to such variables as age, gender, and education. For PROMIS-F, the greater the fatigue, the higher the score reported. Within this index, the mean PROMIS-F score reported with ESF was 49, consistent with previously reported normative values for this scale among people with transtibial amputation.5 By contrast, scores associated with the XF experience were significantly lower (45 points, p = 0.001).2 The notion of reduced fatigue with the XF was further reinforced by a related metric of perceived exertion obtained in the laboratory, discussed below.

TAPES. A final set of self-reported constructs was assessed using the Trinity Amputation and Prosthesis Experience Scales (TAPES), with subscales specific to activity restriction, satisfaction with function, and aesthetics. Subjects reported fewer restrictions on activity and greater satisfaction with their function in the XF condition.2 Satisfaction with aesthetics was identical with the 2 feet—a notable fact, given the unconventional posterior socket mount of XF feet.

Community-observed outcomes

iStockphoto.com 817246860

From experiences gained through patient reporting, all of which favored the XF, discussion moves to community-observed outcomes, represented in this trial by average daily step counts as recorded by activity monitors. Here, the outcome favors ESF by a nonsignificant margin.2 This failure of foot variation to affect average daily step count has been observed fairly consistently in the literature of prosthetics.

For example, in a recent cross-over trial, Wurdeman and co-workers6 observed that, even when ESR feet were compared with the extremely basic and limiting solid-ankle cushioned heel foot, differences in average daily step were insignificant. There are 2 possible explanations for this finding: First, average daily step count is more representative of distances encountered during the course of an individual’s routine than foot function. Second, an individual will execute that routine regardless of his (her) current prosthetic components. Given that the stride length of the sound-side extremity was, on average, 3 cm longer with the XF than those observed with the ESF, the modest decline in steps taken may be partially explained by the roughly 62 additional meters traversed during the course of the mean daily step count of 4109 steps.

Laboratory outcomes

Moving from the community to the laboratory, several additional constructs were observed. Gait speed was recorded in both study conditions, with a nonsignificant increase observed in the XF.2 In a related assessment, the 6-minute walk test, a standardized test of walking endurance, recorded the distance traveled in a 6 minute-interval in both conditions. This metric also favored the XF by a nonsignificant margin.2

The walk test was supplemented by the patient-reported metric of perceived exertion immediately following the test. Here, the difference in favor of the XF was significant, with participants reporting less exertion, on average, in this condition (p = 0.05).2 Considering the construct of walking endurance alongside the related construct of perceived exertion, the researchers noted that nearly one half of participants experienced a clinically significant improvement in their 6-minute walk test or their perceived exertion.2 This collaborative comparison captures the nuanced relationship between performance and exertion on the standardized walking test.

In addition to standardized walking tests, the researchers collected and reported on energy expenditure observed in each of the 2 conditions.3 Subjects were assessed at 3 self-selected walking speeds: slow, comfortable, and fast. As with several of the constructs described above, a nonsignificant decrease in energy expenditure was observed with the XF. However, subsequent analysis determined that different subjects experienced reduced energy consumption with different feet, with trends suggesting which patient types appeared to benefit from each foot type. The 16 participants who performed better in the XF condition walked a longer 6-minute walk test distance and had a greater average daily step count.3 Those with a shorter distance and lower step count did better in the ESF condition.


Just as the concept of the crossover has found its way into the automotive industry, so has it begun to emerge in lower-extremity prosthetics. Observations from a recent cross-over clinical trial have identified those constructs in which XF might prove beneficial, as well as key considerations that might identify patients who will benefit from this emerging technology.

Phil Stevens, MEd, CPO, FAAOP, is an orthotist, prosthetist, and medical writer and editor in Salt Lake City, Utah.

  1. Popely R. Crossovers push sedans down similar path as station wagons. <em>Chicago Tribune.</em> March 3, 2017. www.chicagotribune.com/classified/automotive/sc-crossover-sedans-autocover-0302-20170302-story.html. Accessed July 12, 2018.
  2. Morgan SJ, McDonald CL, Halsne EG, et al. Laboratory- and community-based health outcomes in people with transtibial amputation using crossover and energy-storing prosthetic feet: a randomized crossover trial. <em>PLoS One.</em> 2018;13(2):e0189652.
  3. McDonald CL, Kramer PA, Morgan SJ, Halsne EG, Cheever SM, Hafner BJ. Energy expenditure in people with transtibial amputation walking with crossover and energy storing prosthetic feet: a randomized within-subject study. <em>Gait Posture.</em> 2018;62:349-354.
  4. Miller WC, Deathe AB, Speechley M, Koval J. The influence of falling, fear of falling, and balance confidence on prosthetic mobility and social activity among individuals with a lower extremity amputation. <em>Arch Phys Med Rehabil</em>. 2001;82(9):1238-1244.
  5. Amtmann D, Morgan SJ, Kim J, Hafner BJ. Health-related profiles of people with lower limb loss. <em>Arch Phys Med Rehabil</em>. 2015;96(8):1474-1483.
  6. Wurdeman SR, Schmid KK, Myers SA, Jacobsen AL, Stergiou N. Step activity and 6-minute walk test outcomes when wearing low-activity or high-activity prosthetic feet. <em>Am J Phys Med Rehabil</em>. 2017;96(5):294-300.

Leave a Reply

Your email address will not be published.

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