When Australian podiatrist and researcher Angela Evans, PhD, recently published a suggested clinical pathway for treatment decisions about pediatric flatfoot, it seemed to many readers a reasonable approach that recognized existing evidence as well as the guidelines promulgated by the American College of Foot and Ankle Surgeons. Imagine her surprise, then, when the article provoked a small tempest of outrage in the American podiatric community.
The subsequent exchange of letters to the editor of the Journal of the American Podiatric Medical Association exposed a deep rift in clinicians' attitudes about therapeutic approaches to pediatric pes planus. And although the debate has been waged primarily by podiatrists thus far, the underlying issues have significant implications for any lower extremity practitioner who treats children.
The sticking point: Evans' recommendation that children whose feet are asymptomatic but flatter than expected for their age group should be monitored rather than automatically treated with orthoses.
"It's a common presentation, but opinions about management fall into camps," said Evans, who in addition to her clinical practice is a researcher and lecturer in the division of health sciences at the University of Adelaide. "It ranges from the conservative, never-treat-it camp to the enthusiastic, treat-it-frequently camp. With the renaissance of evidence-based medicine, it behooves us as clinicians to be more scientific and informed about our approach. That's what my proposal was about."
Paul Scherer, DPM, founder of ProLab Orthotics in Napa, CA, and immediate pas chairman of applied biomechanics at the California School of Podiatric Medicine in Oakland, agreed that ideas about treating pediatric flatfoot have swung between extremes.
"There has been a pendulum," Scherer said. "In the fifties, we were treating every child with flatfoot. We'd just come out of the polio epidemic, and braces and special shoes were commonplace. It became an industry, and whenever that happens you overdo it. Then the medical profession overreacted to the fad, and by the seventies and eighties everyone was opposed to treating it. Now we are swinging back to somewhere in the middle of actually identifying which children need treatment and which don't."
As Evans points out in her paper, there is no universally accepted definition of pediatric flatfoot. The ACFAS guidelines note that flatfoot may exist as an isolated pathology or as part of other clinical entities that include generalized ligamentous laxity, neurologic and muscular abnormalities, genetic conditions and syndromes, and collagen disorders.
ACFAS further draws a distinction between rigid flatfoot, which is typically associated with underlying pathology, and flexible flatfoot, characterized by an arch that appears normal when nonweightbearing but that flattens on stance. Scherer refers to the "flexible" condition as hypermobile pediatric flatfoot and defines it as a lowering of the medial longitudinal arch, an eversion of the calcaneus greater than 7°, and a lack of improvement through the age of seven.
All the contention about treatment choices is about flexible flatfoot, however, because clinicians generally agree about the approach to these other conditions. Within the realm of flexible flatfoot, for that matter, there is consensus that symptomatic children should be treated. The controversy relates to those who are asymptomatic, and that turns out to be a lot of kids.
Slight Outcomes Research
The research to date is relatively meager-there are no long-term longitudinal studies of orthotic intervention in pediatric flatfoot-but a few papers shed light on the prevalence of the problem and the parameters of the dispute. For example, one study of 441 subjects found that flat feet of various types are ubiquitous in infants, common in children, and within the normal range in adults. Another paper reported that of 579 schoolchildren (mean age 9), roughly 17% had moderate to severe flexible flatfoot. Researchers at the University of Vermont followed 125 beginning walkers for four years and determined that all normal toddlers had pes planus (primarily flexible) and that arches developed regardless of footwear, though they developed faster if the kids wore shoes with arch supports. Because hyperpronation was present in 78% of five-year-olds, the authors concluded that the condition was normal for the age.
Spanish researchers studied flatfoot prevalence in 1181 schoolchildren ages 4–13, classifying their footprints into three grades of flatfoot. Even though only 2.7% of the children met the authors' diagnostic criteria for flat feet, 14% (168) were receiving orthopedic treatment. Stranger still was that only 28% of the children diagnosed as actually having flat feet were being treated. In other words, within the same cohort both undertreatment and overtreatment were common. A recent study of 835 Austrian children between ages three and six found that flatfoot prevalence was 44%, decreased naturally with age, and was more common in overweight children (a patient demographic with which clinicians are becoming all too familiar) and in boys. The researchers concluded that more than 90% of treatments were unnecessary in the children studied.
Paper after paper, then, finds flatfoot – particularly asymptomatic flatfoot – so common in young children that to consider it a pathology bends our understanding of the term so much that it essentially becomes meaningless. Or does it?
The Clinical Pathway
The Evans paper looked at the three randomized, controlled trials that have studied orthotic intervention in children with flatfoot. One found no significant differences between treatment and control groups. Another found that orthoses significantly improved pain, function, and quality of life in children with juvenile idiopathic arthritis. The third, which like the first looked specifically at flexible flatfoot, reported no significant differences. Evans then extrapolated from the emphasis on diagnostic accuracy in the ACFAS guidelines and proposed a three-stage "traffic light" approach in which symptomatic patients (red) are treated to alleviate symptoms and those with normal asymptomatic flatfoot (green) are left alone.
The yellow light belongs to those who are asymptomatic but "nondevelopmental" (a term that has caused confusion in the podiatric community but that in this context essentially means outside what is usually considered the range of normal development. In the ACFAS guidelines, synonyms are "physiologic" for developmental and "nonphysiologic" for nondevelopmental). According to Evans's pathway, children in this middle or "yellow" category should be monitored and reassessed every few months rather than automatically treated.
This is the recommendation that has caused all the trouble, even though it is very similar to the ACFAS paper, which states that "periodic observation is indicated in non-physiologic flexible flatfoot," and that "children with asymptomatic flexible flatfoot should be monitored clinically for onset of symptoms and signs of progression."
"The contentious part is what we put into the yellow zone, when the feet are flatter than we expect for age but are not overtly painful," Evans explained. "When I graduated from podiatry school in the 1980s, we had been led to believe that if we treated young children early, we could correct the skeletal morphology in their feet. No study is perfect, but the 1989 trial by Dennis Wenger's group showed no structural change in young children over a number of years in four intervention groups versus controls. Those who are enthusiastic to treat want to use orthoses in those children, but we really don't have the evidence to support that. Every clinician needs to be aware of the trap of starting to practice from habit, and the responsibility of remaining current with the emerging scientific literature that may change the way we practice. It's a bitter pill for some practitioners to swallow."
Edwin Harris, DPM, the lead author of the ACFAS guidelines, is a clinical assistant professor of orthopedics and rehabilitation in podiatry at Loyola University's Stritch School of Medicine in Chicago. He agrees with the basic premise of Evans's approach and acknowledges that it provides a practical direction given the lack of long-term research.
"We simply don't have data to help guide us," Harris said. "You have to examine the child and decide whether the approach you take will achieve the goal, and you have to consider the long-term consequences of doing nothing."
In his letter to the editor in the May–June 2009 issue of JAPMA, Joseph D'Amico, DPM, stated that pediatric flatfoot's prevalence "should not be construed as ‘normal' or ‘ideal.' The usual absence of symptomatology in its early stages is an unreliable indicator of optimum foot function." D'Amico is professor and past chairman of the Division of Orthopedic Sciences at the New York College of Podiatric Medicine, and in his private practice in New York City, roughly a quarter of his patients are children.
D'Amico's statement signals the heart of the conflict. He and like-minded colleagues believe that even though nearly all children start out with flat feet, there are many asymptomatic cases that demand attention; otherwise, years down the road, the adult will bear the pain of the child's neglect. We are not, D'Amico says, evolutionarily perfect beings, and our feet bear the brunt of our imperfection.
"Pronation is a poor postural position that sets the stage for future dysfunction, deformity, and disability," D'Amico told LER. "Children have structural imperfections present at birth, including loose ligaments and both osseous and neuromotor immaturity, so that malpositioned bones are subject to the deforming effects of gravity. The keystone of the longitudinal arch is the navicular, and that doesn't even begin to ossify until three and a half years of age."
D'Amico readily acknowledged that most children's feet tighten up naturally as they get older; the problem, he said, is that it's difficult for the clinician to tell which ones will be structurally sound and which won't.
"Whether it hurts or not is irrelevant in a child under age six because most are asymptomatic," D'Amico said. "They don't start hurting until they are 30 or 40 years old. Kids born today can probably expect to live until 90 or 100, and their ability to walk will determine their quality of life when they are older."
It isn't just the future people should worry about, he stressed.
"The problem with the developing foot is that it's too mobile," he said. "The calcaneus is everted and it is not an efficient lever for support, stability, or propulsion."
One recent study seems to support this contention. Researchers in Taiwan performed gait analysis on 377 preschool children aged two to six years and found that those with flexible flatfoot performed physical tasks poorly and walked more slowly versus their non-flatfooted peers. The implication is that there is more to symptomatology than reported pain.
D'Amico advocates treatment as early as seven to nine months of age in some children.
"The earlier treatment begins, the more favorable the prognosis," he said. "If the bones are aligned properly the arch supports itself, and the structure and alignment of the bones in the foot and ankle influence every other part of the skeleton."
D'Amico treats pediatric patients if calcaneal eversion is visible-a threshold he estimates to be roughly 5°.
"You have to look at how much the talus protrudes, how far the forefoot gets abducted, how much bulge there is on the medial segment," he said. "You can have a low-arched foot that is okay because it's just structured low-it's low when the child is seated and when he stands, and the subtalar joint neutral is pretty close to where it should be."
In cases that present less favorably, D'Amico said that when he's been able to treat patients early, their feet form well and they typically don't need further orthotic management by the time they reach their twenties.
Data From the Trenches
D'Amico has been in practice for more than 35 years, and such grizzled clinicians tend to trust the long-term outcomes in their experience more than they trust shorter-term outcomes from studies and clinical pathways based on them. Legitimate questions remain, however, about the statistical support for a cause-and-effect relationship between early intervention and adult outcomes. For one thing, there's a sample bias: practitioners see adult patients in pain, and if a lot of those patients have flat or overly pronated feet, it's natural to draw a conclusion about cause when some evidence suggests the relationship may be merely associative.
In a study of 246 U.S. Army infantry trainees, researchers at Walter Reed found that the 20% with the flattest feet actually had a lower injury risk than those with higher arches. Scientists in Seattle reported that in 99 grocery-store employees studied, there was no correlation between arch configuration and pain scores. And in a study of 2100 military recruits in Saudi Arabia, researchers found that although wearing shoes during childhood, obesity, and urban residence were significantly associated with flatfoot, there was no association with symptoms and that flexible flatfoot was not a cause of disability.
This doesn't necessarily mean that D'Amico and practitioners who agree with his approach only imagine themselves to be preventing problems later in life. Some research suggests that flat feet in adults may be associated with a greater risk of injury or other problems. For example, in an Israeli study of 83 female infantry recruits, lower arches appeared to increase the risk of ankle sprains. An Australian study of 42 participants found that midfoot morphometry predicted successful treatment of patellofemoral pain with orthoses. And another Australian study of 10 subjects with plantar fasciitis and 10 controls found that fascial thickening was associated with lower arch height in those individuals who reported pain.
Such relatively small studies aside, however, the overall body of research does call into question the degree to which flat feet are associated with symptoms-and by implication, the extent to which childhood intervention is appropriate.
Finding Common Ground
D'Amico's colleague at NYCPM, Russell Volpe, DPM, sees room for common ground.
"I don't think the majority of children included in Dr. D'Amico's definition of developmental flatfoot resolve by age seven," he said. "I think Dr. Evans is talking about a much smaller group, those with physiologic flattening as a nonpathological entity. I agree that we can observe and don't need to intervene in those cases."
In the case of non-physiologic (i.e., abnormal) flatfoot , however, Volpe agrees that the podiatrist should seize the opportunity to produce a better outcome.
"How do you decide whether a given case is physiologically normal?" he asked. "There are measurements to help you separate the groups: the degree of eversion of the heel; migration of the talus off of the calcaneus into a planter-flexed and adducted position with bulging of the medial structures; deviation of the forefoot on the rearfoot; malposition of the hallux on weight bearing. These are some of the things I would look for as a clinician, as well as structural comorbidities such as torsions, equinuses, and varus or valgus deformities in the leg or the foot that may be contributing to abnormal pronation."
Volpe evaluates the child's gait to see if it is abnormal or apropulsive, and talks to the parents to see if there are issues with footwear.
"I hear from parents all the time that they have trouble fitting the child because of wide feet," he said. "But often these kids don't have anatomically wide feet; they are excessively pronated. When I take that child back to neutral through my treatment, the foot becomes narrower and starts fitting normal shoes more easily."
Volpe is sensitive to the selection bias inherent in his profession.
"When children come to pediatric foot specialists, the percentage that's going to be in need of treatment is very large," he pointed out. "There's no argument that there's a significant number of children who are just physiologically flat and will be fine by age six without our help. But the kids who come to see me are not those kids. A lot of them are not symptomatic yet, but they're on the track to be, and I don't want to wait around to find out who will be."
Despite this apparent alignment with D'Amico, Volpe nevertheless agreed with much of Evans's article-though he felt the middle "yellow" group could have been better defined.
"It should have been clearer that when the expert finds any number of these factors [e.g., those described above] there is indication to act," he said. "If Dr. Evans had included a more comprehensive clinical discussion of that group, I think she would have squelched some of this reaction. There's a lot of merit to what she says, and she's certainly organized it algorithmically in a very appealing way. But in my opinion she missed the opportunity to be truly successful by glossing over that middle group." (Evans said she is, in fact, in the process of addressing that shortfall.)
Volpe takes greater issue with Evans's treatment recommendations, however, which tended to downplay the importance of custom orthoses.
"That's where she really poked the bear," Volpe said. "A mild case in a younger child can be treated effectively with good prefabricated devices, but as the deformity becomes more moderate to severe, and as the child's age increases, I move quickly to custom devices."
Although Evans and everyone else interviewed for this article decried the paucity of outcomes data, Volpe draws the opposite conclusion from this than Evans does.
"The absence of data is not a reason to go back under the turtle shell and act like we don't have the proof that it works, so we should stop," he said. "We have the three studies she cited, some of which are flawed; the flaws have to do with clearly defining the deformity group and the design of the orthosis. If you put garbage in a shoe and it doesn't work, does that mean a good orthosis won't work either? Those researchers often failed to define and include enough about the devices so readers are comfortable that they used an adequate corrective appliance."
Because the research is so scant, in other words, practiced physicians have to use their experience and judgment too.
"There are all these things we do in our clinical exams when feet are at risk, and that's a lot to guide us even if the outcomes studies don't make it a slam-dunk," Volpe said. "I see child after child who have improved outcomes in their quality of life, whether they had symptoms or not when I started. Activity level, sports participation, style of walking-I see consistent, reliable, and marked improvement."
Scherer agrees with Volpe when it comes to clinical experience.
"We have to make decisions based on what we know-on a combination of evidence-based information and anecdotal information," he said. "And I also agree with Evans that children with hypermobile pediatric flatfoot fall into three categories, and there are those in that yellow category that should be observed. But any notion that all of those children will grow out of the problem has to be dispensed with. One of the cardinal rules of medicine is to do no harm, and a pair of prefabricated orthoses is not going to do any harm."
According to Harris, however, a more conservative approach may be indicated.
"As these kids approach age three, the majority have repaired themselves," he said. "Then there are the stragglers who are flat but asymptomatic. Are we going to try to prevent problems in adulthood that may or may not occur? I don't know that that is a legitimate argument for treatment. I think Evans is on target, and in a large sense she paraphrased what we said in the ACFAS guidelines."
Evans stood her ground while striking a conciliatory tone.
"We have to be bold enough to look at the science-what it says and what it doesn't say," she said. "But that is in no way discounting the value of clinical experience of people who have practiced for a long time, which is always to be used in balance with the science."
Regardless of any other considerations, Scherer added, Evans's paper is a valuable start.
"It may not be the final solution," he said, "but it's important that at least we're looking at a standard for treatment that we can further modify as we gain more evidence."