Correlative Study Between the Sesamoid Bones under the Head of the First Metatarsal and the Development of Hallux Valgus Determined with Radiographs

Objective. To study the correlative between the sesamoid bones under the head of the rst metatarsal and the development of hallux valgus determined with radiographs.Methods.The measurements were performed on the X-ray of 300 normal feet and 300 cases of hallux valgus. The following parameters were measured: hallux valgus angle(HVA); the rst-second intermetatarsal angle(IMA) between the axes of the rst and second metatarsal;the length of the second metatarsal (cid:0) CD (cid:0) ;the position of tibial sesamoid(TSP ) measured the percent formed between the tibial sesamoid and the centreline of the rst metatarsal;the position of bular sesamoid(FSP) measured tangent value between bular sesamoid bone and lateral cortex of rst metatarsal bone ; the absolute distances (AB) from the centre of the tibial sesamoid to the long axis of the second metatarsal, the absolute distances (EF) from the centre of the bular sesamoid to the long axis of the second metatarsal and the absolute distance (GH) from the centre of the tibial sesamoid to the centre of the bular sesamoid. Then calculate the ratio of AB to CD (K1), EF to CD (K2) and GH to CD (K3). Results.HVA moderately positively correlates with TSP and moderately negatively correlates with FSP in subjects with HVA ≥ 20°. HVA and FSP are strongly negatively correlated in the hallux valgus group. Conclusion.The dislocation of sesamoid bone under the rst metatarsal head is an important pathological factor leading to valgus. HVA is positively correlated with TSP and negatively correlated with FSP.


Background
The sesamoid bones on the rst metatarsal head are an important part of the structure and function of the forefoot .It plays an important role in regulating pressure,reducing friction and changing the direction of muscle traction [1][2][3][4] .Whether the sesamoid bones dislocation is corrected or not is an important index about evaluating the curaive effect of valgus correction [5][6][7][8] . Many things are puzzling at the moment, there are a few reports about the quantitative index of the normal position of the sesanoid bones on the rst metatarsal head and clinical cases about the relationship between sesamoid dislocation and hallux valgus.The image research is blank in the aspects of the sesamoid bones at present. In this study,the correlation between sesamoid bones on the rst metatarsal head and hallux valgus was analyzed through the comparative study of normal foot and hallux valgusso, as to provide an important reference for the clinical treatment of hallux valgus and hallux valgus orthosis.
Subjects and Methods

Patients
In hallux valgus groups (the hallux valgus angle (HVA) ≥ 20°and the angle formed between the axes of the rst and second metatarsal (IMA) > 12°),measurements were conducted on 300 feet of 300 subjects (42 feet of 42 men, and 258 feet of 258 women; age 14 to 72 years, average 43.4 ± 2.1 years); the foot with large HVA was conducted when both feet were hallux valgus. Among them, there were 102 patients with family history (102/300, 34.3%), 86 patients with history of trauma (86/300, 28.7%), 30 cases of shoes with unreasonable design (30/300, 10.0%), and 82 cases of other patients (82/300, 27.3%). All patients who visited the hospital were object between 2014 and 2016. On the same period, measurements were conducted on the right foot of 300 subjects in the control group (102 feet of 102 men and 198 feet of 198 women; aged 8 to 74 years, average 43.6 ± 3.2 years). Potential subjects were excluded if they had a history of chronic rheumatoid arthritis, cerebral palsy, cerebral infarction, peripheral nerve paralysis, or other disease factors. None of the subjects had a history of operations for treating hallux valgus; patients had previously either received no treatment or had been treated conservatively.

Inspection Method
The GE De nium 6000 DR lming machine is used to shoot positive and oblique images of the foot. The tube voltage is 60 ~ 65 kV, and the tube current is 100 mA. The radiographs of the feet were taken weightbearing with normal and hallux valgus feet to ensure that the foot is perpendicular to the central axis of the tibia and bula and the rays pass through the back of the foot .

Measurements
The following parameters were measured in the X-ray image of the feet with normal and hallux valgus feet: the angle between the rst metatarsal axis and the proximal hallux phalanx axis (the hallux valgus angle, HVA); the angle formed (IMA) between the axes of the rst and second metatarsal; the length of the second metatarsal (CD); the absolute distances (AB)from the centre of the tibial sesamoid to the long axis of the second metatarsal, the absolute distances (EF) from the centre of the bular sesamoid to the long axis of the second metatarsal and the absolute distance (GH) from the centre of the tibial sesamoid to the centre of the bular sesamoid. Relative to the position of the tibial sesamoid to the centreline of the rst metatarsal, the percent formed (TSP) between the tibial sesamoid and the centreline of the rst metatarsal, and de ned by the position of the tibial sesamoid located on the medial of the centreline of the rst metatarsal and not to exceed the centreline of the rst metatarsal is 0%; the centreline of tibial sesamoid located on the centreline of the rst metatarsal axis established is 50%, located on the right of the centreline of the rst metatarsal and completely exceeding the centreline is 100%; others are between 0-100% relative to the position of the bular sesamoid to the lateral of the rst metatarsal. The percent formed (FSP) between the bular sesamoid and the rst metatarsal, and de ned by the position of the bular sesamoid located on the of the lateral of the rst metatarsal and not connected to the rst metatarsal is 0%. The bular sesamoid and the rst metatarsal completely coincide with 100% (Figs. 1 and 2). Next, the relative distances, de ned as the ratio of these absolute distances to the length of the second metatarsal, were calculated to adjust for foot size. The ratio of AB to CD is K1; EF to CD is K2; GH to CD is K3.

Statistics
SPSS 20.0 statistical analysis software was used. First, the measurement data were expressed as ±s, and two independent sample t-test was used for comparison between the two groups. TSP, FSP, K1, K2 and K3 are independent variables, HVA and IMA are dependent variables, and multiple linear regression analysis was used in all patient measurements (in the control group and the hallux valgus group) during the study period. We investigated the correlation between the TSP, FSP, K1, K2, K3 and HVA and IMA. The variables with signi cant differences in the above regression analysis were the dependent variables. Parameter (P < 0.05 was considered signi cant) analysis was performed as independent variables, HVA and IMA are dependent variables using the multiple linear regression analysis to investigate the correlation in the control group with HVA ≥ 20 and in the hallux valgus group (HVA ≥ 20°and IMA > 12). |r|≥0.8 was considered strongly correlated; 0.4≤|r|<0.8 was considered moderately correlated; |r|<0.4 was considered weakly or not correlated. P < 0.05 was considered signi cant.

The incidence of hallux valgus
Hallux valgus is a common disease in foot and ankle surgery, which is a kind of foot compound deformity in which the metatarsal toe tilts outward and exceeds the normal physiological angle [9][10][11][12] . It is characterized by the adduction of the rst metatarsal bone, hallux valgus deformity and subluxation of the rst metatarsophalangeal joint in severe cases [13] . It can be accompanied by other foot deformities such as cross toe, hammer toe, at foot and so on [14][15] . The incidence of hallux valgus was reported in literature as 2%-50% [16] with a completely difference. In 1990, the National Institutes of Health calculated that the incidence of hallux valgus was 5.1% [17] . There are more females than males, and the ratio of male to female is 1: 4.63 [18][19][20] . In our study, the average age of hallux valgus patients was (43.4 ± 2.1) years old, the youngest patient age was 14 years old, and the male to female ratio was 1: 6.14 (42 / 258). At present, the pathogenesis of hallux valgus is not clear, which is closely related to congenital and acquired factors [21][22] . The common factors are heredity and trauma [23] . Hallux valgus is autosomal dominant hereditary disease [24] . In this group, 102 patients (102 / 300) had family genetic history, and the age of onset was generally after 30 years old, while those without genetic history were generally younger, generally before 30 years old. It was reported in previous literature that irrationally designed shoes were the second leading cause of hallux valgus, while trauma (86 / 300, 28.7%) was the second leading factor besides heredity in 300 patients with hallux valgus. However, irrationally designed shoes were only found in 30 cases (30/300, 10.0%). Therefore, sesamoid dislocation may be the main factor leading to hallux valgus after trauma.

Correlation between tibial sesamoid bone and hallux valgus
It was reported [25] that the following relationships between the sesamoids on the head of the rst metatarsal and hallux valgus. The sesamoids on the head of the rst metatarsal are in dynamic balance because of pulling by two group muscles. The inside end of the abductor hallucis muscle and the exor hallucis brevis muscle is the tibial sesamoid. The outside end of the abductor hallucis muscle and the exor hallucis brevis muscle is the bular sesamoid. When the dynamic balance of the two groups of muscles is broken, it weakens the strength of the abductor hallucis muscle or increases the strength of the exor hallucis brevis muscle. All of these factors can cause the occurrence of hallux valgus, and when the HVA exceeds 30 ~ 35°, the abductor hallucis tendon is displaced plantarward, and the adductor hallucis, exor hallucis brevis, and abductor hallucis rotates in a lateral direction. This usually leads to pronation of the ankle, and the abductor muscles inside the exion axis of the rst metatarsophalangeal joint normally moved towards the ankle. Laterally, the adductor muscles that lost the epicondylar abductor muscles further pull the metatarsal to create eversion, pulling the medial capsular ligament, especially the osseous capsule of the joint capsule, and making the medial metatarsophalangeal joint capsule tense through the "capstan mechanism". The pulling force on the inside of the rst metatarsal head is generated, and at the same time, the lateral side of the humeral head is pushed and the rst metatarsal inversion is aggravated, resulting in relative sesamoid subluxation. The majority of the tibial sesamoid in normal foot are located on the inside of the central longitudinal axis of the rst metatarsal shaft [26] . Hardy and Clapham [13] reported the seven-position scale for sesamoid displacement from the AP view. They reported that a correlation was found between the stage of the position of the sesamoid and the hallux valgus angles. Yildirim [14] reported the four-position scale for sesamoid displacement from the axis view. Judgement standard was the relationship between the position and the bony crest of the rst metatarsal. That is, the tibial sesamoid bone did not exceed the bony crest of the rst metatarsal at 0 degrees, less than 50% is 1 degree, more than 50% are 2 degrees and completely exceeding the bony crest of the rst metatarsal is 3 degrees. This method can explain the subluxation of the sesamoid. The results of Aper RL, Saltzm an CL, B row n TD [10] showed that there were different degrees of sesamoid dislocation in hallux valgus, and the more obvious the dislocation of sesamoid bone was, the larger the HVA and IMA were (P < 0.05). There was a signi cant correlation between sesamoid dislocation and HVA. The correlation number between sesamoid dislocation and HVA was 0. 47 (P < 0.05). The correlation coe cient with IMA was 0.43 (P < 0.05). There was no correlation between the distance from the tibial sesamoid bone to the second metatarsal axis and the ratio of the second metatarsal length K1 (P > 0.05).
In this study, through the measurement and statistics of the normal foot, it was found that there was no signi cant correlation between HVA and TSP (P = 0.0445). IMA was only correlated with K1 (T = 3.37, P = 0.001) ( Table 5). The results are not consistent with the study of Aper RL, Saltzm an CL, B row n TD. Aper RL only measured the K1, while this study measured the AB, GH, K1, and K3. HVA was correlated with K1 and K3(T=-3.273 P = 0.002 and T=-2.025 P = 0.046 ) in the normal foot. The results show that TSP was moderately correlated with HVA and IMA in subjects with HVA > 20° (|r|=0.494, P < 0.05). TSP, FSP and HVA and IMA are strongly positively correlated in the hallux valgus group (|r|=0.824, P < 0.05). In other words, the more intersecting of the central axis of the tibial sesamoid bone and the rst metatarsal bone, the larger the valgus angle (HVA). When the HVA > 30 °, the tibial sesamoid bone tends to deviate to the lateral side of the central axis of the rst metatarsal bone. This is similar to AperRL,SaltzmanCL,BrownTD. Through measurement and statistics of the normal foot group, we found that the normal position of the tibial sesamoid bone (TSP) is 17.8%-34.3% (intersection values between the tibial sesamoid bone on the head of the rst metatarsal and the central longitudinal axis of the rst metatarsal shaft), with an average of (28.2 ± 2.1)%. At present, the normal position of TSP has not been reported at home and abroad. The discovery of this quantitative index lays a foundation for future study of sesamoid lesion on the rst metatarsal head. 3.22 Correlation between bular sesamoid bone and hallux valgus When the relationship between the sesamoid bones and the rst metatarsal head changes for various reasons, the dynamic fulcrum of the sesamoid bones movement shifts, gradually forming the metatarsal seed joint and causing fatigue and strength changes of the surrounding muscles or tendons (the strength of the lateral muscle group increases and the strength of the medial muscle group weakens) [26][27][28] . So the metatarsal toe gradually tilts to the outside, which makes a normal line of force change [29] .
While the metatarsal head tries its best to move medially in order to maintain the support of the foot. When the rst metatarsal head is pulled by the toe seed ligament and other soft tissues, the rst metatarsal head is pronated, resulting in valgus deformity [30] . Another theory holds that the abductor muscle and adductor muscle and the exor brevis muscle form a joint stop, part of which ends on the tibial sesamoid bone. The original balance of the muscle strength of the metatarsal bone promotes the metatarsal rotation during the relative displacement of the sesamoid bone. The sesamoids on the head of the rst metatarsal are in dynamic balance because of pulling by two group muscles [31] . So in theory, either sesamoid dislocation could lead to hallux valgus [32][33] .In this study, it was found that the normal position of sesamoid bone (FSP) intersected with the lateral cortex of the rst metatarsal bone accounted for (52.3 ± 3.2) % of the total area of sesamoid bones. The purpose of this study was to nd a quantitative index for judging the dislocation of sesamoid bones on the rst metatarsal head. Among all the subjects in this group, HVA was correlated with FSP, K2 and K3.  Approval from the Institutional Review Board was obtained and in keeping with the policies for a retrospective review, informed consent was not required." Consent for publicationNot applicableAvailability of data and materialData sharing not applicable to this article as no data-sets were generated or analyzed during the current study. If you do not wish to publicly share your data, please write: "Please contact author for data requests."Competing interestsThe authors declare that they have no competing interests.FundingNot applicableAuthors' contributionsXiaozhong Li carried out the measurements studies, drafted the manuscript. Dongxue Liu participated in the design of the study and performed the statistical analysis. Xufang Wang conceived of the study, and participated in its design and coordination and helped to draft the manuscript. All authors read and approved the nal manuscript. Corresponding authorCorrespondence to Xiaozhong Li. Acknowledgements Not applicableAuthors' information Xiaozhong Li  Figure 1 the position of the tibial sesamoid is 50%, the position of the bular sesamoid is 50%.