Perhaps the easiest way to relate HV to a Cartesian “matter in motion” model is by a subjective description of the deformity as a “bump” on the medial and/or dorsal-medial aspect of the first metatarsal-phalangeal joint. Although we generally lack extensive or definitive epidemiologic studies on the deformity, any physician with any degree of clinical experience can relate that this “bump” represents a common patient complaint. Even the term “bunion ” itself derives from the Latin word “bunio” translating to turnip [8, 60], and admittedly it is not difficult to imagine an inflamed first metatarsal-phalangeal joint as the red and relatively bulbous vegetable.

However, most descriptions of the deformity simply as an abnormal “bump” of bone are historical. In 1856 Volkmann described the deformity as an “exostosis” on the first metatarsal head which in turn pushed the hallux into a lateral direction [61]. Mann and DuVries have presented the possibility of a congenitally wide first metatarsal head when considering etiology, which they proposed might lead to chronic inflammation, resultant capsular thickening, periosteal reaction, and further osseous enlargement [60]. Schoenhaus and Cohen specifically described a dorsal-medial “hypertrophy” of bone associated with the deformity [62]. The idea of the deformity as an abnormal hypertrophic bump is even propagated with some contemporary educational tools (Fig. 3.1). But relatively simple logic tells us that if HV was primarily an abnormal osseous growth on the first metatarsal head, then successful surgical correction of the deformity would merely involve resection of the exostosis. One might also assume that if this were indeed the case, then we would have no need for the dozens of described translational metatarsal osteotomies described for correction of the deformity or perhaps even for this entire textbook on the deformity!

A better description of HV might be relatively normal bone in an abnormal position, in this case the first metatarsal head. In other words, it is more likely that there is not a structural abnormality of the first metatarsal head in HV, but instead the natural progression of the deformity results in it being in a more prominent position because of medial translation/rotation of the first metatarsal as a whole. This concept of an abnormally positioned first metatarsal and hallux as opposed to abnormal new bone growth has been generally accepted for some time [54, 63–66]. Even Dr. David Silver, who somewhat ironically gave his name to the surgical procedure describing the removal of the hypertrophic osseous medial eminence from the first metatarsal, described HV as primarily being positional with very little “actual bone hypertrophy” in 1923 [66].

Two comparative clinical studies have attempted to provide evidence on this topic by quantifying the size of the first metatarsal head and any medial hypertrophy in the HV deformity [3, 67]. These studies objectified the so-called medial eminence by measuring the width of that portion of the first metatarsal head that protrudes out past the normal medial shaft of the first metatarsal. Thordarson and Krewer first measured the width of the entire first metatarsal head and then the width of the medial eminence in a group of 50 ft with HV and compared this to a group of age-matched controls without deformity [67]. They found a difference in the width of the medial eminence of only 0.2 mm between groups. Lenz et al. replicated this study design and found that a group complaining of bunion pain with radiographic HV deformity had both a larger medial eminence and wider first metatarsal head, but by only 1.12 mm and 2.81 mm respectfully [3].

These studies had several limitations that should be appreciated by critical readers, however. The precision of the utilized measuring instrument was unclear, and there was no attempt to correlate fractions of millimeters of deformity with clinical symptoms. Further, they relied on completely transverse plane radiographic evaluation which would not be expected to take into account frontal plane rotational positional changes. The frontal plane position of the first metatarsal has been demonstrated to contribute to perceived deformity on the dorsal-plantar radiographic projection [68].

All this is not to say that some morphological changes do not occur at the joint during the progression of the deformity. In 1887 a pathologist named Dr. W. Arbuthnot Lane provided a detailed description of the atrophic and degenerative changes that were likely to occur in the setting of long-standing HV as the hallux proximal phalanx base moved in a relative lateral direction on the head of the first metatarsal. Pertinent to our discussion on etiology, he cautioned that “the surgical pathologist has but too frequently mistaken the cause for the effect” [64], indicating that the visible and often symptomatic changes seen to the first metatarsal head in HV more likely represent the end result of the disease process and not necessarily the initiating cause.

Even if an accumulation of hypertrophic bone is not present on the first metatarsal as a primary cause of the deformity, some portion of the medial first metatarsal is typically excised and/or remodeled during surgical correction [9]. Thordarson and Krewer found that the width of the medial eminence decreased in size by a mean of 2.7 mm following HV surgical correction, for example [67]. Wen et al. performed hematoxylin and eosin stains of 123 resected medial eminences in patients undergoing HV surgery and noted consistent changes of “extensive chronic inflammation” [69]. In reviewing the pathology reports from 315 consecutive resected medial eminences from HV surgery, Oh et al. also found that the overwhelming majority (97.5%) were found to have degenerative changes and that it was not cost-effective to routinely send these specimens for specific histopathologic examination [70]. Similarly, Prasitdumrong et al. [71] and Uchiyama et al. [45] performed histologic analyses of the medial collateral ligament from feet with HV and found evidence of degenerative changes, abnormal mechanical properties, and collagen fiber differences. Bone cyst and bursa formation are also common at the first metatarsal-phalangeal joint. Further, Haas described a loss of trabeculation from the medial first metatarsal in the deformity consistent with long-standing disuse atrophy [65].

Degenerative arthritic findings have also been noted in the first metatarsal-phalangeal joint in the setting of HV deformity [72–76]. Mafart observed the presence of some minor osteophytic changes positively associated with age in a group of first metatarsals with HV deformity exhumed from a French necropolis [72]. Roukis et al. prospectively evaluated a series of 166 feet undergoing HV corrective surgery and found evidence of degenerative changes in a large majority, particularly along the inferior and medial aspects of the first metatarsal head [74]. Doty et al. similarly found intra-articular degenerative chondral changes on the first metatarsal head positively associated with HV severity [73]. And at least two studies have demonstrated an association between HV deformity and degenerative changes of the sesamoids [75, 76].

When this evidence is taken together, it seems reasonable to conclude that any “bump” associated with the HV deformity, both in terms of clinical appearance/symptomatology and morphological changes, represents the effect of the deformity and not the primary cause. The consistently observed changes to the joint structure, including articular cartilage remodeling, also likely represent mechanical adaptation, chronic inflammatory effects, and degenerative alterations .

Although shoegear has been widely recognized as a potential cause for the development of HV, the evidence in support of this is often circumstantial and unfortunately more often seems to represent the result of recycled historical literature reviews rather than independent critical analyses. Some sources point to investigations which are reported to indicate the lack of HV in patient populations that do not regularly wear shoes; however, these conclusions do not hold up to a critical review of the original sources. Although Sim-Fook and Hodgson did report a hallux valgus rate of only 1.9% in a non-shoe wearing Chinese population, they also reported a rate of metatarsus primus varus of 24.3% and of metatarsus hypermobility of 13.1% [77]. Barnicott and Hardy noted “valgus deviation” of the hallux of barefooted Nigerians which “falls within the European symptomatic range” [78]. In a study of Solomon Islands natives in 1939, James reported that the hallux was “occasionally abducted” [79]. Kato and Watanabe concluded an association between HV and shoegear in part because they subjectively did not observe the presence of HV in a series of Japanese footprints believed to be over 2000 years old [57]. And Engle and Morton’s study of natives in the Belgian Congo actually contains no specific mention whatsoever as to either the presence or absence of hallux deformity, although it does provide several figures indicating that it was observed [80]. These studies seem to indicate the presence of structural first metatarsal-phalangeal joint pathology in populations not exposed to regular shoegear or at the very least do not provide evidence to the contrary. In other words, it certainly seems possible to develop first metatarsal-phalangeal joint deformity in the complete absence of shoes.

It is perhaps more likely that shoegear has a tendency to exacerbate subjective patient symptoms and possibly even exaggerate the progression of the deformity rather than be an initiating cause of structural abnormality. From an evidence-based perspective, a prospective, long-term, longitudinal study rigidly controlling for shoegear and activity behaviors would be required to provide practical data on the effect of a certain type of shoe on deformity progression. This is simply not a practical investigational design, and so available evidence is somewhat indirect and at risk for bias. In 1965, Shine was able to provide data on a unique patient population on the island of St. Helena where regular shoe wear had only recently become commonplace [81]. He found that the percentage of the population with hallux deviation (defined clinically as lateral deviation of the hallux away from the medial border of the foot) steadily increased with the number of years that shoes were worn. However, this study was unable to account for participant age, and understandably those wearing shoes for a longer period of time would be expected to be older. More recently, Klein et al. drew an association between clinical abduction of the hallux with improperly sized shoes in a group of Austrian school children [82]. However, this study was not associated with any subjective symptoms, clinical outcome measures, or an evaluation of deformity progression. It might also be negatively biased as nearly 90% of children were judged to be in improperly fit shoes, somewhat limiting a “control” comparison.

Most of the data that is available on this topic is based only on patient-reported symptoms and surveys of previous behaviors. One might argue that this puts the information at risk for an error in logic exemplified by the Latin phrase “post hoc, ergo propter hoc” or “after it therefore because of it.” In other words, if a patient develops a symptomatic HV deformity, then they might be more likely to retrospectively associate it with an external variable such as shoegear. For example, Munteanu et al. completed a study on monozygotic and dizygotic twins and found patient perception of HV severity was associated with a reported history of consistently wearing constrictive toe box shoes [83]. Borchgrevink et al. noted an association between a reported history of wearing high-heeled shoes and patient symptoms, but did not identify any radiographic deformity differences in a limited controlled cohort investigation [84]. And Menz et al. found that women aged 50–89 years who self-reported hallux valgus pain were more likely to report a history of wearing constrictive footwear between the ages of 20 and 39 [85].

Although a critical analysis of the literature makes it difficult to draw definitive conclusions about the effect of specific shoegear behavior patterns on the development and progression of objective structural first metatarsal-phalangeal joint pathology , it is easier to conclude an association with subjective patient-reported symptoms. And while this is important to a discussion on the etiology of HV, it might arguably be more important to a discussion on patient expectations following HV surgical correction. Several studies have indicated that the ability of patients to comfortably wear their preferred shoegear postoperatively is strongly associated with their satisfaction of the procedure [4, 86, 87].

It is not difficult to cite sources reporting a family history of foot deformity in a majority of patients presenting with HV [20, 56, 88–92]. Few, however, have attempted to determine a specific mode of inheritance. Ola Johnston is typically credited with the first description of HV as being transmitted with a pattern of autosomal dominance with incomplete penetrance, although this conclusion was based on the pedigree of a single male college student in Texas [93]. Perhaps the most detailed investigators were Piqué-Vidal et al. who had 350 participants with HV complete a family history questionnaire extending back three generations [94]. Ninety percent had a history of at least one affected family member, and the authors concurred with a pattern compatible with autosomal dominance with incomplete penetrance. Coughlin and others have further provided some evidence for the possibility of a high relative rate of maternal transmission [89, 90]. It seems reasonable to conclude that HV has a hereditary component.

If there is evidence of a hereditary component to the HV deformity, then a reasonable follow-up question inquires what exactly is passed down. This relates to a potential irony associated with the theory of evolution in that there is a tendency within human nature to assume that the process has reached a stable apex. Although there is certainly evidence that the structure and function of the foot has changed substantially since human beings transitioned to consistent bipedal ambulation, it is in fact likely that this is still a relative work in progress [95–105]. Considering lower extremity structure and function as a dynamic, somewhat variable, and evolving process instead of one which is universal and static is an interesting way to approach the etiology and development of the HV deformity. And while much of what has been specifically described on this topic is owed to the work of Morton, Lapidus, and Hansen [15, 49, 106–108], one might also argue that the strong associations more recently described between both hypermobility and the flatfoot deformity with HV are at least indirectly if not directly related to this discussion as well.

Morton, Lapidus, and Hansen focused on several consistent findings involving the first metatarsal labeled as “atavistic” indicating a relation to ancestral function [15, 49, 106–108]. These include the obliquity of the first metatarsal-medial cuneiform articulation , length of the first metatarsal segment, and equinus of the Achilles complex. To discuss these topics, and really anything found more within the area of biomechanical theory, a divergence from reliance on high levels of clinical evidence and evidence-based critical analysis is required. Readers must also make several assumptions with respect to lower extremity function that admittedly cannot be definitively established and might be open to other interpretations. The most basic of these is that the human foot has generally evolved from more of a grasping function in an arboreal species to a propulsive function in an erect and bipedal species.