In 1925, Geist et al. [18] proposed the simple resection of the accessory navicular bone. He reported that the simple resection of the accessory navicular bone and trimming the residual bony process did not require complete cutting of the posterior tibial tendon(PTT) insertion and thus, could preserve and maintain the continuity and function of PTT. He emphasized the importance of partial PTT freeing and repairing the tendon damaged by stripping. In 1929, Kidner et al.reported using AN resection and PTT reconstruction for the treatment of AN patients in order to maintain the medial longitudinal arch of the foot and the dynamic basis of flexion and inversion of the toes. He suggested separating PTT with small bone fragments, removing the AN and suturing with chromic suture to reconstruct the free PTT with small bone fragments on the medial plantar surface of the navicular tubercle so as to achieve tendon reconstruction and obtain a good therapeutic effect [15]. Macnicol et al. [19] compared the clinical effect of Kidner surgery and simple excision in treating painful AN. He performed Kidner surgery on 26 AN patients with severe flat feet and used simple excision in 21 patients with painful AN. The follow-up results showed no significant difference in foot pain relief between the two groups after the operation (P > 0.05). In their study, Cha et al. [20] performed Kidner surgery and simple resection on 50 AN children, respectively, and 25 cases in each group were followed up for 3 years. The clinical results were good in both groups, and the re-alignment of the foot arch was similar.

However, both procedures have disadvantages. The simple resection of the AN cannot achieve reconstruction and suture fixation of the loose PTT but only surgical removal of the AN. The pain was reported to worsen in patients with severe flat foot deformity, who complained of unrelieved symptoms after surgery, while even progressive aggravation and recurrence were possible [21]. Kidner surgery may cause damage to the medial support system of the foot. If a firm insertion of the PTT cannot be achieved, tendon relaxation is likely to occur. Repeated friction of PTT causes aseptic inflammation around the insertion and local scar hyperplasia, thus preventing pain relief [22]. Choi et al. [23] reported recurrent pain in 9 patients after Kidner surgery. They argued that after Kidner surgery was performed in patients with flat feet, the tension in the attachment point of the PTT increased, resulting in tendon degeneration and recurrent foot pain.

Along with the development and progress of orthopedic implants, many scholars modified the Kidner procedure by inserting suture anchors into the navicular and then reconstructing PTT using the suture anchors. Dawson conducted a comparative study of 13 patients, 7 of whom were treated with suture anchors, finding that the use of anchors could shorten the postoperative recovery time of walking [23]. Kakihana et al. performed PTT reconstruction with thread anchors in 15 adolescent patients who were followed up for at least 1 year, finding significant relief of symptoms during short and mid-term follow-up with satisfactory radiology results [25]. Many studies have shown that using suture anchors could make operations simple, minimally invasive, reliable, and safe. Moreover, it could help to avoid secondary surgery. The suture anchor has excellent tissue compatibility while in contact with bone tissue. The built-in strong wear-resistant and tension-resistant polyester fiber suture can stably and firmly weave and suture the PTT, thus providing stability and benefiting PTT fused to the navicular. Therefore, we sutured and fixed PTT with suture anchors in both groups in the present study.

The relationship between AN and flatfoot has always been the focus of interest among researchers. Kidner argued that the abnormal anatomy of PTT weakens the maintenance force on the medial longitudinal arch of the foot, resulting in flatfoot. At the same time, the alignment of PTT changes due to the protruding AN, which may cause impingement from AN, resulting in painful bursitis, PTT tendonitis, and eventually flatfoot [15]. Kiter et al. pointed out that the PTT inserts directly on the AN could weaken the stability of the talonavicular ligament. As PTT has no supinator function without its distal attachments, the gastrocnemius-soleus complex acts at the talonavicular joint, making the passive structures of the longitudinal arc collapse and causing flatfoot [26]. Bernaerts et al. suggested that the presence of type II and type III AN could easily lead to posterior tibial tendinopathy because, in the presence of type II and III AN, most of the posterior tibial tendons are inserted into the AN, shortening the insertion point of the PTT, reducing the leverage of the ankle on the posterior tibial tendon and resulting in increased tendon stress [27]. In the current study, 1 patient from the MK group had pain in the AN resection area. After wearing footpads for 6 months, the pain was slightly relieved. The patient underwent a subtalar joint stabilizer (HyProCure) procedure, and the pain disappeared after the second surgery, which indicated that the pain of AN is different from flatfoot.

Some scholars reported that AN does not cause flatfoot and that AN and flatfoot are accidental phenomena. Sullivan et al. suggested that the presence of the AN does not participate in developing flatfoot [28]. Micheli and colleagues argued that the AN does not cause flatfoot. On the contrary, the excessive tension and traction of the PTT during the ossification of the navicular induce the formation of AN [29].

Senses et al. [30] performed AN simple resection on 8 patients with flat foot deformity and restored PTT's continuity, finding that PTT had insertion points on both AN and navicular. After resected the AN, they sutured the main trunk of PTT to the navicular. During the 2-year follow-up, all 8 patients could complete the single-foot heel raise. There was no change in the lateral Meary angle of 6 patients, while the angle was only reduced by 2° in 2 patients. They suggested that this procedure did not significantly improve the collapse of the longitudinal arch of the foot, thus providing evidence that the AN was not associated with flatfoot. In the present study, we stripped the PTT from the outer layer of the accessory navicular cartilage surface in all 23 cases, which certified that PTT has insertion points both on AN and navicular. Our results also showed that the continuity of the PTT could be kept intact and does not need to be cut off. Therefore, we proposed to preserve PTT insertion and reconstruct the PTT by folding it twice and suturing it to the navicular. By using a double fold, the posterior tibial tendon was tightened, and the overlapping distance of the tendon could be determined intraoperatively according to the relaxation of PTT. This method retains the advantages of the Geist and Kidner procedures and helps to avoid the anchor displacement caused by the excessive concentration of stress points on the anchor after PTT amputation and reconstruction. The improvement of the AOFAS score after the operation suggested that our methods could effectively stop foot pain and restore foot function. The improvement in Meary angle and Pitch angle suggested that the alignment of the foot was better after PTT reconstruction.

During the follow-up period in the current study, there was only 1 patient in the FS group who experienced pain in the insertion area of the posterior tibial tendon caused by increased tension of the posterior tibial tendon in the right foot. No reoperation was performed for pain and movement disorder symptoms; the follow-up is still underway. It is possible that the maintained angle of the ankle joint plantar flexion and forefoot varus was excessive when we sutured the PTT, resulting in an over-shortening of the PTT, which caused tension increase. This highlights the critical importance of maintaining the position of the foot and ankle when folding PTT and suturing it to the navicular. PTT should be sutured to the medial wall of the navicular when the ankle joint is maintained in a neutral position. Unfortunately, we only had a short-term follow-up results. Yet, we plan to observe the clinical results further during mid-term and long-term follow-up, aiming to obtain more detailed data.