Fig. 6.1
(a) Mapping of the CORA for first metatarsal transverse plane deviation component of HAV deformity. The metatarsal has no intrinsic angular deformity but is deviated from its normal orientation relative to other first ray components. Although this may not represent the true mechanical axis definition for the first ray, it represents the anatomic axis definition of the deformity and the site of correction to realign the first metatarsal. (b) The degree of IMA does not change the actual level of the deformity and in reality does not define the deformity despite the common convention of assigning procedure choice based on IMA
The next important consideration when describing the anatomy of the deformity is understanding the individual planar components of the deformity. Typically surgeons rely on an anterior-posterior (AP) radiograph almost entirely to define the deformity by measuring the intermetatarsal angle (IMA) , hallux valgus angle (HVA) , tibial sesamoid position (TSP) , and the joint surface angle known both as distal metatarsal articular angle (DMAA) and proximal articular set angle (PASA) . It must be pointed out that these are all two-dimensional observations which define only the transverse plane components of the deformity. To identify and characterize the other planar components of the deformity (frontal and sagittal), we must look at different landmarks, and it is very helpful to look at the anatomy on axial radiographic projection as well as the lateral radiographic view. Though it’s also important to understand that since the AP radiograph is a two-dimensional projection of the three-dimensional anatomy, an out-of-plane deformation, such as frontal plane rotation of the first metatarsal, can substantially change several visible cues on the AP radiograph, and we will discuss the effect rotation has on each of the common radiographic findings.
The practice of preferentially considering the transverse plane of the deformity by relying primarily on AP radiographic measurements gives an incomplete understanding of the deformity and in our opinion is one of the main factors driving poor outcomes and recurrence. If we analyze the majority of the most popular osteotomy procedures, it is clear that correction priority is in a single plane (transverse) with most procedures either angulating or sliding the first metatarsal in the transverse plane while failing to address either the frontal or sagittal planes to a meaningful degree. Despite the published description of the frontal plane component of the first ray deformity dating to the 1950s [35], it has not been common to address this component of the deformity in a bunion operation. Recently there is a renewed interest in the frontal plane position of the first metatarsal and sesamoid alignment , and there are many current publications illustrating the effect frontal plane rotation has on common paradigms of preoperative bunion evaluation and the selection of the corrective procedure. In these studies frontal plane rotation has consistently been observed to be in the direction of eversion (valgus or pronation are equivalent) and has a significant and dramatic effect on the alignment of the first MTPJ including the sesamoids (Fig. 6.2).
Fig. 6.2
(a) The AP radiograph shows typical transverse plane angular findings commonly used for decision-making regarding procedure choice (IMA, HVA, TSP). This view shows a TSP of V. (b) Semi-weight-bearing sesamoid axial view clearly showing the sesamoids in their normal anatomic location medial and lateral to the crista. Frontal plane eversion of the first metatarsal relative to the plane of the lesser metatarsals gives the appearance of sesamoid subluxation on the AP view when the sesamoids are in reality in normal alignment relative to the metatarsal head
Scranton and Rutkowski [47] presented a series of sesamoid axial radiographs to observe the position of the metatarsal. They found feet with bunions had a mean of 14.5° of metatarsal pronation (valgus orientation), while normal feet had a mean of 3.1° of valgus metatarsal orientation. They concluded that the three structural deformities present in a bunion must be corrected: the abducted hallux, the adducted metatarsal, and the pronated or valgus metatarsal position. Mortier et al. [36] also used sesamoid axial radiographs to observe the position of the metatarsal in a bunion deformity. Their novel method of both patient position and measurement showed a mean of 12.7° of metatarsal pronation in feet with bunion deformities. They concluded this rotation was due to metatarsal cuneiform instability rather than torsion of the metatarsal shaft and that valgus metatarsal rotation in bunion deformities is systematic. Eustace et al. [13] devised a way to measure pronation of the first metatarsal based on the observation of the location of the inferior proximal tuberosity of the first metatarsal base. The lateral translation of the tuberosity that takes place with metatarsal pronation or valgus position was established in a cadaveric study. They found that the degree of first metatarsal pronation has a linear relationship to the amount of medial deviation of the first metatarsal. They concluded that derotational surgical procedures should be further explored (Fig. 6.3).
Fig. 6.3
AP and sesamoid axial views of three patients. (a) Normal alignment of bone segments used to diagnose HAV and corresponding normal frontal plane rotation of the first metatarsal. (b, c) Patients with HAV showing the AP and sesamoid axial alignment. Note the easily visible eversion of the first metatarsal in the frontal plane relative to the lesser metatarsal plane
Recent computed tomography studies have clarified the position of the first metatarsal in the frontal plane then in normal and bunion feet. Collan et al. [3] first reported on the use of weight-bearing 3-D CT on hallux valgus patients and found that pronation (valgus position) of the first metatarsal and proximal phalanx existed in all ten patients with hallux valgus. While not found to be statistically significant, they found that the amount of first metatarsal rotation of the hallux valgus group was 8° everted versus the control group of 2°. They found that the cuneiform was rotated into valgus to a greater degree than the first metatarsal although they were both rotated. One methodological issue that may confuse their findings is the fact that while the scans were taken weight bearing, the patient was in single leg stance, not in functional angle and base of gait. This fact alters the overall kinematic relationships because in single leg stance, the weight-bearing extremity is externally rotated inducing supination of the foot. Kim et al. [25] evaluated 166 ft with hallux valgus versus 19 normal control feet utilizing semi-weight-bearing 3-D CT analysis and measured the amount of first metatarsal rotation, which they referred to as the α angle. This angle, representing first metatarsal pronation, averaged 21.9° in their hallux valgus group versus 13.8° in the control group. They concluded that the first metatarsal pronation in subjects without hallux valgus is typically less than 15.8°, and that pronation higher than 15.8° is abnormal (Fig. 6.4). Kim and colleagues [25] further identified four groups based on the presence of either pronation of the first ray (87.3% of patients) and/or subluxation of the sesamoid (71.7% of patients) (Fig. 6.5). We also have data to suggest that in a foot without HAV, the first metatarsal and/or the first ray are neither pronated or supinated. Lamo-Espinosa et al. [29] found that in normal subjects, the CT appearance of the sesamoid complex showed no subluxation and minimal metatarsal rotation. The utilization of computerized tomography will provide further three-dimensional information to help elucidate the pathomechanics of hallux valgus. A more detailed analysis of CT studies is discussed in Chap. 5.
Fig. 6.4
Weight-bearing CT scan views of three patients with hallux valgus deformity (a) Patient with small increase in IMA with minimal to no eversion of the first metatarsals. (b) Patient with moderate increase in IMA with notable eversion of the first metatarsal relative to the lesser metatarsals. (c) Patient with a large increase in IMA again clearly showing eversion of the first metatarsal relative to the lesser metatarsal plane
Fig. 6.5
Two patients with HAV and eversion of the first metatarsal relative to the plane of the lesser metatarsals. (a) No sesamoid subluxation from the normal position medial and lateral to the crista. (b) Everted first metatarsal starting to sublux medially off of the sesamoids with the medial sesamoid now partially on the crista
It is clear that a bunion is in reality a triplane deformity with components in the transverse, sagittal, and frontal planes. Despite this anatomical fact, the most commonly accepted paradigm for the correction of a bunion employs transverse plane metatarsal and, to a lesser extent, hallux osteotomies to reposition the metatarsal in the transverse plane only. Osteotomies must almost universally be combined with lateral capsular release and medial plication to reposition the sesamoids under the metatarsal head which cannot be achieved with osteotomy alone in most cases. The common practice of transverse plane metatarsal osteotomy does not fully address the deformity, and it is not performed at the CORA (which is proximal to the metatarsal), both of which are believed to be principal factors in the high recurrence rates that have been identified with metatarsal osteotomy as discussed in the next chapter (Fig. 6.6).
Fig. 6.6
(a) Patient who had a sliding osteotomy without correction of frontal plane eversion. The sesamoids are not aligned in the sagittal plane, and therefore the forces exerted by pull of the long and short flexors on the everted sesamoids are angular and pull the hallux into valgus and abduction. There is a medial force exerted by the hallux on the first metatarsal driving increased IMA and recurrence. (b) In the normal state the movement of the sesamoids and the hallux are predominantly in the sagittal plane without abnormal angular forces induced by metatarsal and sesamoid rotation
At this time, we do not know the exact anatomic site of the frontal plane rotation . That is, whether it is occurring at the TMTJ or at a site proximal. Most likely it is occurring at a combination of joints similar to sagittal plane mobility of the medial column which is well known to occur at multiple joints along first ray including the naviculocuneiform joints, talonavicular joint, and to a lesser extent the TMTJ. Studies by Johnson and Christensen [20] and Dullaert et al. [11] provide insights into both the frontal plane position of the medial column and the mobility present. Using different models both groups showed in a weight-bearing foot, activation of the peroneus longus tendon pulls the first ray into eversion. Dullaert et al. [11] further stated that if this frontal plane mobility was not controlled through TMTJ fusion for correction of HAV that there is a concern for persistent frontal plane deformity. This phenomenon is a potential cause for poor results and recurrence as discussed in Chap. 7.
Effect of Rotation on Sesamoid Position
From our observations and from available literature, it is clear that radiographic tibial sesamoid position can largely be influenced by metatarsal frontal plane rotation rather than solely an observation of the metatarsal moving off of the sesamoids in the transverse plane [1, 6, 7, 19, 48]. In reality, both frontal plane rotation and transverse plane deviation of the first metatarsal produce the positional components of the bunion deformity. Two-dimensional radiographic findings are directly influenced by the three-dimensional deformity.
Several studies have demonstrated a correlation between the degree of sesamoid displacement observed on AP radiographs and the transverse plane severity of the bunion deformity [22, 34]. Discussion of this correlation often includes the observation that there is a constant position of the sesamoids in relationship to the second metatarsal [16, 17, 22, 42, 46] as well as the proximal phalanx to the second metatarsal [26]. The constant relationship of the sesamoid position in the transverse plane lends itself to a proposed process where the first metatarsal slides medially off of a stable and stationary sesamoid apparatus that is tethered in place via ligamentous and tendon attachments. However, it is important to understand that the appearance of the sesamoids on AP radiograph is not always indicative of their actual position in relation to the median crista and the bisection of the metatarsal shaft through the median crista. Frontal plane rotation of the first metatarsal can significantly alter what is seen on the AP radiographic projection. The pronated or valgus position of the metatarsal can give the false appearance that the metatarsal head has migrated off of the sesamoid complex and that the fibular sesamoid resides in the interspace when in many cases the sesamoids are still positioned correctly medial and lateral to the median crista of the rotated plantar first metatarsal head (Figs. 6.3 and 6.4).
Inman [19] used a combination of models and radiographs to show that in a valgus or pronated metatarsal position, the sesamoids appear to deviate laterally in an AP radiograph. However, the comparison of sesamoid axial radiographs to their AP counterparts show the sesamoids are still found in their anatomic positions (in their grooves and separated by the median crista) despite their appearance of lateral translocation. Boberg and Judge [1] make the same observation after bunion correction without interspace release. In the majority of their cases, the preoperative AP radiographs showed apparent deviation of the sesamoids, and the sesamoid axial failed to confirm the sesamoid displacement. They explained that the apparent subluxation of the sesamoids is due to an oblique rotation of the metatarsal head much the way that a medial oblique radiograph shifts the perspective making structures appear more lateral. The authors called into question the use of AP radiographic sesamoid measurement as a tool of bunion assessment. Talbot and Saltzman [48] came to a similar conclusion regarding the use of AP radiographs to evaluate sesamoid subluxation. They found that sesamoid position as estimated from AP radiographs did not correlate to the actual sesamoid position when viewed using a tangential view, a term synonymous with sesamoid axial. The difference between the observations could not be accounted for by changes in MTPJ positioning while obtaining the sesamoid axial view. Because of the valgus (pronated) position of the metatarsal, measurement models based on AP radiographs are not valid in assessing true sesamoid position. These studies are corroborated by a cadaveric study by Dayton et al. [6], in which the first TMTJ was freed and the metatarsal was moved into various degrees of inversion and eversion. With eversion (pronation) of the metatarsal, there was the appearance of lateral displacement of the sesamoids on AP radiograph. With inversion (supination) the apparent sesamoid position was corrected. In this study, the metatarsal clearly did not move off of the sesamoid apparatus, rather rotation altered what was observed on AP radiographs.
Because they recognized the difficulty in assessing sesamoid position from an AP radiograph, Kuwano et al. [28] devised a measurement used to observe sesamoid position on tangential or axial radiographs. Not only did they find a correlation to the degree of HAV and the valgus (pronated) position of the sesamoid apparatus, but they also found the AP assessment of sesamoid subluxation was inadequate to assess true sesamoid position. These results also support the observations from Dayton et al. [6], DiDomenico et al. [9], and Mizuno et al. [35] that varus (supination) rotation imparts correction of sesamoid position on AP radiographs when the coronal plane valgus (pronated) position of the metatarsal is addressed. Kim et al. [25] identified both rotation of the first metatarsal and sesamoid subluxation on CT scans of HAV patients. Both states can exist in isolation and in combination. The striking finding is that, in many cases, the AP radiographic views do not accurately define the position of the sesamoids and thus AP x-rays cannot be reliably used to identify sesamoid subluxation. Obtaining axial views of the sesamoid complex is a necessary and vital part of evaluation and management of the complex triplane deformity of HAV. Similarly, Katsui et al. [24] found a direct correlation of sesamoid displacement with increased severity of hallux valgus and arthritic changes.
If the pronated or valgus metatarsal is a consistent reason for perceived deviation of the sesamoids, what is really taking place with transverse plane translational osteotomies that produce the appearance of restored sesamoid position in AP radiographs immediately post procedure? In the case of a sliding osteotomy that corrects the IMA but cannot produce inversion (supination) rotation to correct frontal plane position of the metatarsal, we hypothesize that iatrogenic subluxation of the sesamoids medial to the median crista creates the perception that the sesamoids are correctly positioned under the metatarsal on the AP radiograph. This occurs after the lateral release and during the medial capsular plication. An additional explanation is that in some cases a degree of frontal plane correction takes place spontaneously when retrograde buckling forces of the hallux acting on the metatarsal are relieved. If the appearance of sesamoid correction is a result of iatrogenic medial subluxation, then the position on AP radiograph would not be maintained over time. The sesamoids would appear corrected on the postoperative film due solely to the lateral soft tissue release and medial soft tissue plication, but over the ensuing months, the sesamoids would find themselves returning to their anatomic position in the sesamoidal grooves, which are still rotated in a valgus (pronated) orientation. This lateral drift, which is in reality resumption of normal position relative to the metatarsal head, is due to the plantar soft tissues including the short and long flexor tendons resuming their linear orientation after joint motion resumes and therefore pulling the sesamoids back to their anatomic location under the metatarsal head, which is still in a rotated position. This sesamoid position relative to an everted metatarsal would mean recurrence of a displaced appearance of sesamoids on an AP radiograph. Though immediately postop sesamoid position would be predictable and within control of the surgeon via soft tissue balancing, long-term maintenance of this position would not be predictable nor under control of the surgeon as the pathologic position of the metatarsal causing the appearance of subluxation has not been addressed. This would also produce deforming forces from the hallux proximal to the metatarsal because of the lateral position of the sesamoids and tendons as described by Mortier (2012) and can result in recurrence of both the HAV and increased IMA (Fig. 6.7). Thus, if frontal plane metatarsal pronation is present as a component of the deformity (PVB Class 2A or 2B, described in Chap. 5), it must be addressed by the corrective procedure to achieve full anatomical correction of the metatarsal sesamoid complex (Fig. 6.8).
