Clinical Outcomes of Single Excision Versus Kidner Procedure for Type II Accessory Navicular Associated with Flatfoot in Adults: Does accessory navicular induce exible atfoot?

with (AN) the


Introduction
The accessory navicular (AN) is a common accessory ossicle of the foot. Type II is most popular type of AN problem. Once the patients with AN became symptomatic, they mostly complained of mid-foot pain and bony eminence, which occurred recurrently after sports or sprain. However, some of them was recognized with a exible atfoot (FFF) at their rst weightbearing radiographic examination. Whether the AN induced in FFF is controversial [1]. The AN is reported up to 21% of population, which are found in patients who have at foot (19%) [2]. Some doctors believe the posterior tibial tendon (PTT) disfunction in this type of FFF. However, without PTT interruption, AN osteosynthesis was reported to gain good outcome [3]. But few articles have reported the rate of FFF after osteosynthesis. We believe that single osteosynthesis may not resolve all biomechanical problems of AN with FFF.
As well known, PTT and various medial ligaments, were noted to be dynamic and static contributors [4].
The AN involved in FFF as an anatomic variant. When the AN present, the PTT stops at the nal insertion in the internal cuneiform and lesser metatarsals instead of at the plantar surface of the navicular tubercle. Instead of pulling directly from the turbucle of navicular, the PTT contracts the lower end of AN and multiple insertions at tarsus as an adductor rather than as a supinator. In very earlier research, Kidner [5] described the term as prehallux, due to the theory that it is a degenerated evolutionary remnant and a progressive evolutionary effort of nature to re-enforce a weakening and pronating foot arch. The relationship between AN and PTT may be essential to reconstruct of medial arch, and how to treat PTT resumption after resection determines the principal theme of this paper. After single excision of AN without PTT reconstruction, foot arch was not collapsed. It is proved that the PTT was not only resources of maintain the arch. We designed this study to compare the clinical changes in FFF after AN resection and analyze the contribution of AN to FFF.

Materials And Methods
The study included all adult patients from January 2014 to January 2020, who originally complained the pain and bony prominence at the medially to the navicular tubercle. 61 adult patients were con rmed type II AN with atfoot by radiographic examination and separated into two group randomly. Group A consisted of 30 patients who underwent single AN resection, and Group B consisted of 31 patients who underwent the Kidner procedure with an anchor. The exclusion criteria were patients had other types of AN but type II, a history of a local steroid injection, local infection, or underlying diseases such as uncontrolled diabetic mellitus, rheumatoid arthritis, or seronegative spondyloarthropathy.
All patients was palpated before surgery at medial side of foot, especially at the synchondrosis, posterior edge of medial prominence, along and insertion of PTT course.
Prior to the surgical treatment and at the nal follow-up, clinical evaluations were made in all patients using the American Orthopedic Foot and Ankle Society (AOFAS) mid-foot score, visual analog scale (VAS) for pain, and radiographic assessments performed by anteroposterior (AP) and lateral radiographs.
Other data were designed for this research (Fig. 1A-C). The points that de ne the inclination of PTT are described as follows: the posterior border of posterior malleolus is taken as point P; the tangent point at the scaphoid tubercle or end of AN is point E; the lowest end of AN is point N; the base of the rst metatarsal bone is point M; the PTT declination angle (PDA) is formed by the lines PE and MN, which is measured as the projection line of PTT at lateral view. The reason for choosing the rst metatarsal base is considered for the enormous changes after the operation. The AN-N joint inclination angle (ANJCA) is formed by the joint line (SQ) and the weight bearing surface. The AN size includes the length and height in the lateral view and the length and width in the AP view.

Surgical Technique
An incision is made along the course of PTT, with its center at the AN. The AN was found to carry the main attachment of the PTT. With a thin osteotome and a rongeur, the prominence, which includes the whole AN and part of the corresponding amount of the navicular, was removed. Then, an additional naviculoplasty was performed(9).
In Group A, AN resection and naviculoplasty were performed, the brocartilaginous margin was sutured to the engulfed ligamenum by 2-0 Ethibond polyester sutures (Ethicon, New Brunswick, NJ, USA). In Group B, the tendon was inferiosuperiorly transplanted by an anchor (DePuty, New Brunswick, NJ, USA) so that it could heal with the fresh stump of the tarsal scaphoid. The position of the anchor is located at the center of the surface and direction points at the Cuboid/Metatarsal 4 joint. It is strengthened by sutures through the adjacent ligamentous tissues, xed with the foot in mild supination. The tibiospring ligament or soft tissue was strengthened to the PTT by 2-0 Ethibond polyester sutures (Ethicon, New Brunswick, NJ, USA). Finally, the wound was routinely closed.

Postoperative management
The patient was maintained for the rst two weeks. The rehabilitation began at the third week, including physiotherapy and splinting, with gradual return to weight bearing in a shoe until six weeks. The patients started full weight bearing and returned to normal life at three months.

Statistical analysis
All statistical analysis was performed using SPSS software (version 12.0; SPSS, Chicago, IL). A chisquare test and one-way analysis of variance (ANOVA) was performed to assess the difference of baseline characteristics between the group A and group B. A one-way analysis of variance (ANOVA) was performed to assess the outcome of clinical and radiologic in the preoperative and postoperative between group A and group B. A P value less than 0.05 was considered signi cant.

Results
All patients underwent surgery by three senior doctors in a single institution. In such cases, patients whose lateral column lengthening (LCL) osteotomy was added to the correct forefoot adduction were excluded. Two cases in Group A and three cases in Group B were combined LCL procedures. In total, 56 patients (56 feet) were included in this study. The mean follow-up period was 22.29 and 20.86 months (range: 6-60 months). Informed consent was obtained from all patients. The demographics of the patients are shown in Table 1.  20.86(6-60) χ 2 = 0.725 Some ANs were so large, but only palpation was located at the inner ends of the navicular without other symptoms. The normal leverage of the PTT was maintained, and there were mild at feet. For these patients, palpation was at the synchondrosis and bony prominence; for the others, whose foot collapsed markedly, the pain was at the inferior edge of the insertion, not at the course of PTT.
Hypertrophic AN inferomedially extends the scaphoid tubercle, and PTT distributes complexity more than normal. The principal bers of PTT stop on the AN, partial bers bypass through the dorsal mid-foot, and the slender medial part reverses and joins into the tibionavicular ligament, which strengthens the medial capsule of the TN joint (Fig. 2). In all observed cases where the AN was dissected off the considerable size, the PTT was displaced forward by its attachment to the stump of the navicular.
Radiographic and clinical results were separately tabulated for each group (Tables 2 and 3).  According to the pre-operative view, patients with ANJCA above 70 degrees described the location of pain at the lower edge of the AN, and the others with ANJCA below 70 degrees described the location at the AN-N joint and bony prominence. While the bypassed bers were scattered to the plantar tarus, the gross PTT extended the AN and pulled the arch indirectly inward and upward. The impingement between the AN-N joint leads to the intensity of AN in MRI in patients with ANJCA below 70 degrees. We hypothesize that the reason is that the contraction of PTT is erected to the AN-N joint. The tenderness of the PTT engulfs was obvious in the other patients with a larger ANJCA. We believed that the force of PTT did not affect the AN-N joint, especially the dorsal side. The force arm of the PTT increased after detouring, which was easily damaged. The size of the hypertrophic AN was de ned in this study as a length of more than 1.5 cm in either lateral view or AP view, with an ANJCA angle inclined to less than 70 degrees. Meanwhile, in the group with AN less than 1.5 cm and a larger ANJCA angle over 70 degrees, the FFF was severe. We believe that the theory in this type is more likely caused by dorsal ligament dysfunction.
After the AN resection and naviculoplasty, the PDA was signi cantly increased. The inferior partial tubercle of the navicular was removed, and the PTT was re-stumped navicular or reconnected to the resumption. The radiographic indices of the midfoot signi cantly decreased in both groups, and the forefoot and hindfoot data signi cantly changed in Group B. According to anatomy [9], the middle component and tarsometatarsal component of the PTT inserts on the cuneiforms, cuboid and peroneal canal. In Group A, PTT forced itself on the forefoot. The medial arch was reconstructed in Group B because the midfoot was directly forced by PTT reattachment. The hindfoot was not signi cantly changed in either group.
Two groups of patients were satis ed with the pain release, gaining increasing AOFAS midfoot scores and decreasing VAS scores. However, all patients in Group A felt that the strength decreased after 3 months and improved after 6 months.
Fore patients in Group B felt popping at the level of the medial malleolus while running at 6 months. These patients have larger AN than 2.0 cm. The patient whose AN was less than 2.0 cm did not feel popping after the operation. Scar pain could be provoked by resisting the action of the PTT or heavily tapping the incision.

Discussion
AN has been recognized in patients who suffered FFF, with increasing frequency. However, the relationship between AN and FFF remains controversial [11]. PTT dysfunction is widely accepted as a signi cant contributor to this deformity. Chronic tension and shear forces by the PTT at increasing pronation have been implicated as the etiologic biomechanical mechanism [12][13]. The abnormal anatomy of FFF typically starts at the PTT, but dysfunction in this tendon by itself is not enough to cause substantial deformity. The cumulative damage of remaining structures causes the typical mal-alignment of FFF. We believe that the uique anatomy around AN is the alternative reason for FFF, instead of an indirect factor.
According to the literature, the broad insertion of PTT that engulfs the tuberosity of the navicular and reaches the rst cuneiform similar to a cuff [14]. In type II, the principal dynamic stabilizer of the longitudinal medial arch, which is the posterior tibialis muscle, is attached to the AN instead of directly at the keystone of the triple arch complex. This engulfed part can be de ned as a ligament, i.e., accessory navicular cuneiform ligament (ANCL), which is the brous connective tissue that connects the AN and the medial cuneiform. We hypothesize that the relationship between ANCL and PTT is identical to that between patellar ligament and quadriceps tendon [15].
Like spring ligament, ANCL gradually strain causes FFF. Therefore, the pain of FFF derived from AN, the pain locates at inferior of AN-N joint rather than the course of PTT. This leads to the decreasing strength in Group A. However, they cannot retain their original shape when extended beyond their characteristically viscoelastic for a prolonged period of time. The ANCL becomes prone to future injury. Finally, atfoot occurs with forefoot adduction. The patients in this study had got high AOFAS-midfoot score because there was no strength evaluation of AOFAS scores and all patients paid attention to the rehabilitation postoperatively.
Park's radiologic results showed that the hindfoot was more in equinus, the midfoot was more pronated and abducted, and the forefoot was more abducted and pronated in patients with AN than the normal control group [16]. We agreed with it that the suspend force at arch was medialized. Therefore, PTT reconstruction could correct at deformity. The midfoot indicates, such as naviculocuboid overlap and talonavicular coverage angle, has decreasing more signi ciently in Group B.
The various surgical options are used to treat the atfoot with AN, including simple resection [17]; extended excision with reconstruction of PTT [18][19]; percutaneous drilling [20]; and fusion between the supernumerary bone and the main navicular [21][22]. However, these treatment is still controversial. Many doctors believe AN xed on the navicular by removing the cartilage parts between them. The function of PTT was backed at the keystone. However, the xation has a lot of disadvantage. First, the xer irritates the attachment of PTT where it passes through [3,23], moreover, it has been an approximately 20% rate of nonunion [21]. Chung [24]was reported that near 20% (7/34) patients was not satis ed the results. Second, the expanded tubercle could impact the dorsal TN joint anteriomedially. Tabionavicular part of the deltoid ligament attaches at the dorsal side of the AN, where is crowding of soft tissue, leads to discomfort at the position of supration. Once the atfoot formed gradually, this part of ligament was felt complaint by hypertension at pronation of foot. And the function of PTT is impaired by the close approach of the AN to the medial malleolus.
With AN, PTT forces harder to maintain the medial arch, due to the major bers insert at the navicular indirectly. At the lateral view, PDA increased signi cantly after operation, the effect of PTT on the arch upward has been induced by the angulation of pull and the shorter distance through which the PTT produced. The inward contraction approach more lineally, which it increases the su ciency of PTT.
There is no doult without de ciency. First, the naviculoplasty may have an effect on the result of PDA, naviculocuboid overlap at lateral view and talonavicolar coverage angle at AP view. Second, the AOFAS midfoot score has no part to evaluate the strength, PROMIS evaluation may be the alternative. Third, we had measured the length of shortening of PTT, however, it was abandoned because the measurement at the AP view may have an error due to foot supination/pronation.
The FFF with AN may be induced by AN and its synchondrosis. The weakened plantar ligament of synchondrosis was impaired under chronic tension and shear forces may be implicated as the etiologic biomechanical mechanism. AN excision or it with PTT reconstruction could release the pain and bene t the PTT pulling su ciency.

Declarations
Phenotypes and anthocyanin content (" DAP 30 " means 30 days after pegging; " DAP 50 " means 50 days after pegging ). a Seeds of pink parental line (YZ9102), red parental line (ZH12) and F4 lines with pink and red testa in different development stages, b relative anthocyanins content in parental lines and homozygous F4 lines with testa color character.

Figure 2
Distribution of candidate SNPs and InDels per chromosome. a Candidate SNPs and InDels using ΔSNP algorithm with a cutoff of ΔSNP > 0.5. b The top 1% SNPs and InDels using ED algorithm. c Distribution of candidate SNPs/InDels on Chr.12. The signi cant region identi ed for red testa pheotype is shaded (109.9-117.7 Mb).

Figure 4
Gene structure and expression analysis of candidate gene. a Gene structure of candidate gene and the locations of the SNPs. b Sequence alignment of the ANR in different species. c Expression of the candidate gene in the parental lines and pools with the red and pink testa, respectively.  Genotyping results of SNP:Ch12.117288744 by KASP(Kompetitive Allele Speci c PCR).a The scatter plot with axes x and y represents allelic discrimination of this site genotypes. The red, green and blue dots represent the mutant homozygous, heterozygous and wild-type homozygous, respectively. b Validation of diagnostic marker in YZZH12 and YZZH2 populations.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.