History

The roots of posterior tibial tendon (PTT) insufficiency are found in early writings on tenosynovitis. In 1818, Velpeau described the first case of noninfectious tenosynovitis in the hand. In 1895, de Quervain described stenosis-related tendovaginitis in the first compartment of the dorsal carpal ligament of the wrist in more than 900 patients. Kulowski,¹⁰⁵ in 1936, was the first to publish tenosynovitis of the sheath of the posterior tibial tendon, documenting one case. More than a decade later, the first large series of PTT tenosynovitis was published by Lipscomb at the Mayo clinic.¹¹³ Lapidus and Seidenstein¹⁰⁹ commented on two cases of PTT tenosynovitis in 1950, stating, “nonspecific chronic tenosynovitis must be considered a rarity, particularly at the ankle.” These early published reports underestimate its prevalence.

In 1955, A.W. Fowler¹⁰⁷ reported on seven cases of PTT tenosynovitis. He found “the condition was seldom diagnosed early, and it was often mistaken for osteoarthritis of the ankle and treated conservatively without relief. At operation, the tendon sheath was swollen and thickened and the tendon greatly enlarged. The inflamed synovium was excised, with relief in all cases.” Although anecdotal, this was the first documented series in which tenosynovectomy was used and found to be successful. This concept was reinforced by Langenskiöld,¹⁰⁷ who, in 1967, documented six cases of PTT tenosynovitis treated surgically after failure of conservative management. These patients experienced great pain relief after debridement of the proliferative granulation tissue.

Key,⁹⁸ in 1953, documented the first case of PTT rupture (partial). In the prelude, he states that the case should be of interest, showing great foresight. This worker’s compensation case documented the classic signs and symptoms of the condition, including the not-so-unusual missed early diagnosis. This partial rupture was treated with excision of the torn component and debridement of the residual thickened tendon, leaving the patient with 15% disability at the ankle.

Griffiths,⁶⁷ in 1965, recognized the difficulty in correcting deformity after late reconstruction of PTT rupture, but all of his “spontaneous” ruptures were in patients with rheumatoid arthritis. Thus confounding factors obviate the direct correlation between PTT rupture and deformity correction. Sixteen years after Key’s article, Kettelkamp and Alexander⁹⁷ explored spontaneous PTT rupture in four patients without systemic disease. Missed in this article was the prevalence of spontaneous rupture because the authors begin by stating that this mechanism is a rarity. However, the authors pinpoint the challenge of correcting deformity created by delay in treatment. Three of the four patients were treated operatively without correction of tendon length (the retracted tendon was bridged with extensor digitorum longus tendon graft or filled with a Z-plasty lengthening), leaving the patients with residual pain and no correction of deformity. Results at best were rated fair.

In 1974, Goldner⁶¹ revisited the topic of progressive talipes equinovalgus in traumatic and degenerative conditions of the posterior tibial tendon. This article studied nine patients with either of these etiologies, all resulting in the common pathway of an acquired flatfoot deformity. The authors correctly recognized a disorder of the medial plantar calcaneonavicular ligament resulting in limited support through elongation resulting from repetitive stress. Surgical intervention was done for deformity correction as well as pain relief. Transfer of the flexor hallucis longus to substitute for the deficient or absent PTT was done in all cases. The tendon was sutured to the periosteum under the navicular. This tendon was chosen over the flexor digitorum longus (FDL) because of its “more tendinous” structure, larger muscle mass, and the potential ability to elevate the sustentaculum tali to combat hindfoot valgus. In addition, the medial plantar calcaneonavicular ligament was plicated, after experience in not doing so resulted in persistent or progressive deformity. The authors also found failure in simply advancing or plicating the posterior tibial tendon. They also believed that a contracted gastrocnemius–soleus complex contributes to the deformity and must be addressed at the time of surgery.

Eight years later, in 1982, the concept of spontaneous rupture of the PTT resurfaced with a scientific presentation by Mann and Specht.¹²² The authors reviewed eight patients undergoing a variation on Goldner’s flexor hallucis longus (FHL) tendon transfer by using the FDL tendon as the source replacement tendon. Rationale for avoiding the use of the FHL tendon for transfer centered on the importance of maintaining full flexion strength in the hallux in compromised patients. This paper will be reviewed more thoroughly in the surgical section of this chapter.

After the preceding article, the history of PTT dysfunction has followed the standard course of newly recognized syndromes in the history of medicine: a search for etiology, pathophysiology, and successful treatment options among multiple investigators. The ideas presented by these investigators will unfold in the following sections.

Anatomy and Function

The substance of a tendon is formed by a group of fasciculi. An epitenon surrounds the tendon and contains an outer synovial layer. Next, a paratenon, made of loose areolar tissue, surrounds the epitenon, carrying the tendinous structures. The next layer comprises the synovial sheath. The synovial sheath is usually continuous with the epitenon, forming a mesotenon that acts as a vascular channel for the tendon.⁷ The mesotenon of the PTT sheath is not continuous. This anatomic variant will derive clinical significance as the potential for vascular insult is explored.

The posterior tibialis muscle arises from the interosseous membrane and adjacent surfaces of the tibia and fibula in the proximal one third of the leg. The myotendinous junction appears in the distal one third of the leg. The PTT courses directly behind the medial malleolus at a relatively acute angle. The groove is shallow, and the flexor retinaculum binds the tendon tightly into this groove. Thus the tendon passes posterior to the axis of the tibiotalar joint and medial to the axis of the subtalar joint, plantar flexing and inverting the hindfoot.¹⁷⁴ In fact, the PTT is located farther medially from the axis of the subtalar joint than any other tendon about the ankle, and it therefore has the greatest degree of leverage to bring about inversion of the subtalar joint. It then passes beneath the calcaneonavicular ligament to insert into the tuberosity of the navicular. It is unique in that it has eleven insertion domains: the navicular, the sustentaculum tali, the medial, middle, and lateral cuneiforms, the cuboid, and the bases of the second, third, and fourth metatarsals.^74,174 By its insertion into the midfoot, the tibialis posterior both adducts and supinates the forefoot.

Tendons possessing synovial sheaths have altered directional courses or are bound by tunnels or retinacula. Linear tendons (e.g., the Achilles tendon) often do not have sheaths. Tendons that contain a synovial sheath are generally located at the distal portions of the upper and lower extremities.¹⁸⁶ The sheath consists of three layers: a parietal layer lining the deep fibrous surface or fibroosseous canal, a visceral layer covering the tendon, and a mesotenon connecting the visceral and parietal layers serving as one source of the tendon’s blood supply. The PTT is unique in that it does not contain a complete mesotenon and must receive its vascular supply through other channels. Synovial sheaths act to decrease the frictional forces encountered during tendon motion.¹⁸⁶ As gliding occurs, the visceral layer glides against the parietal layer. The mean length of the tendon sheath is 71 mm in men and 66 mm in women. The tendon sheath runs approximately 45 mm proximal to the apex of the medial malleolus and continues approximately 26 mm distal to the peak of the malleolus.¹⁷⁰ The excursion of the PTT is only 2 cm.

The PTT receives its blood supply from four regions: the vessels proximal to the muscle insertion, the connective tissue peritendinous arterial network, the arteries running to the tendon in the triangular vincula, and vessels from the periosteal insertion of the tendon.¹⁷⁰ Frey et al⁵⁴ found that the tendon receives its blood supply at the musculotendinous junction via the posterior tibial artery. They noted that a mesotenon was present in the PTT proximally, providing an additional network of vascular channels from the posterior tibial artery. The visceral layer also provides additional proximal blood supply, using this mesotenon as a conduit. This visceral layer remains closely adherent to the epitenon proximally. Distally, at the tendon–bone interface, the periosteal vessels provide the tendon’s blood supply. These periosteal vessels are terminal segments for both the medial plantar branch of the posterior tibial artery, supplemented two thirds of the time by the medial tarsal artery, a branch of the dorsalis pedis artery.

The PTT functions to stabilize the hindfoot against valgus forces or eversion. The tibialis posterior is a stance phase muscle, firing from heel strike to shortly after heel lift-off.⁸⁹ It decelerates subtalar joint pronation after heel contact through eccentric contraction.⁷ At midstance, it stabilizes the midtarsal joints. During the propulsive phase of stance, the tibialis posterior adducts the transverse tarsal joint, initiating inversion of the subtalar joint. This action has two beneficial effects on the gastrocnemius–soleus complex: it locks the transverse tarsal joint, allowing the gastrocnemius–soleus complex to maximize the plantar- flexion force during gait, and it shifts the direction of pull of the Achilles tendon further medially, allowing the gastrocnemius–soleus complex to become the primary invertor of the subtalar joint through increased leverage. In doing so, the foot can become a rigid lever that supports the propulsive phase of gait.

Quantifying this action with respect to gait, during the normal walking cycle, eversion occurs in the subtalar joint at the time of initial ground contact. The tibialis posterior becomes functional at about 7% of the cycle, the soleus muscle at about 10% of the cycle, and the lateral head of the gastrocnemius muscle at about 25% of the gait cycle. Dynamic electromyography suggests that this initial eccentric contraction of the posterior tibial muscle lasts from 7% to 30% of the gait cycle during stance. After this, progressive inversion occurs, starting at approximately 30% of the cycle through concentric contraction of the tibialis posterior.161,175 The tibialis posterior muscle is silent during swing phase, where its antagonists enjoy maximal benefit. The primary antagonist of the posterior tibial muscle is the peroneus brevis muscle, functioning to abduct the midfoot and evert the hindfoot. Cross-sectional area studies note that the peroneus brevis muscle is 41% as strong as the posterior tibial muscle (relative strength of the PTT is 6.4, and of the peroneus brevis, 2.6).¹⁶⁵ Physiologically, this is manifested through the primary function of the peroneus brevis in unlocking the transverse tarsal joints and everting the hindfoot during the non–weight-bearing swing phase of gait.

Finally, stabilization of the longitudinal arch by the PTT remains a topic of debate. Most authors agree, however, that there are both static and dynamic forces at work. Static support theorists fall into two camps, those believing the foot acts as a truss and those believing the foot acts as a beam. The truss theory is supported by Lapidus.¹⁰⁸ A truss works by creating two struts that meet at an apex, supported at the base by a tie rod, thus forming a triangle. As the apex is loaded, compressive forces are applied to the struts, and tensile forces are applied to the tie rod. As long as the tie rod remains intact, the struts do not collapse, and the truss holds firm. Relating this model to the anatomy of the foot, the tie rod becomes the plantar aponeurosis. Hicks⁷⁵ believes that this model becomes critical at toe-off, when the windlass mechanism has maximal effect.

The beam theory, proposed by Sarrafian,¹⁶¹ supports a less rigid construct. In this model, the foot is a curved beam that sags when it is loaded. Forces generated at the midportion of the beam are compressive on the convex side of the beam and tensile on the concave side. The curved portion of the beam consists of the bones of the midtarsus. Thus tension directly affects the structures on the concave side of these bones, namely, the plantar ligaments. Anatomically, these ligaments consist of the long and short plantar ligaments, the calcaneonavicular (spring) ligament, and the bifurcate ligament. All are attachment sites for the posterior tibial tendon.

The spring ligament consists two distinct bands: a stronger superior medial calcaneonavicular ligament and the inferior calcaneonavicular ligament just lateral to it. The latter structure has a medial plantar oblique and the plantar inferior band.

Dynamic support of the arch revolves around the tibialis posterior muscle and the intrinsic musculature of the foot. Support of the arch by the intrinsics is discussed in the next section. With respect to the posterior tibialis muscle’s contribution, Kapandji⁹² suggests that contraction of the tibialis posterior adducts and plantar flexes the navicular on the talar head. In doing so, it buttresses the medial longitudinal arch against collapse. In addition, the ligamentous attachments of the PTT have an effect by pulling the cuboid medially along with the navicular through the bifurcate ligament. This cuboid then pulls the calcaneus medially through the strong calcaneocuboid ligament, providing additional support to the talar head through the anterior and middle facets.

Pathophysiology of Posterior Tibial Tendon Rupture

The PTT has a limited excursion of only 2 cm. Thus any insult, no matter how minor, that lengthens this tendon has an adverse effect on its function. This lengthening may be gradual or acute, depending on the underlying pathologic process. The inversion power provided by the PTT has been underestimated by some, the thought process being that all of the posterior compartment muscles act to provide this function. Jahss⁸⁵ noted that the normal inversion power of all posterior calf muscles combined is 12 to 15 pounds of torque. Patients with PTT rupture undergo a substantial reduction in this force, lowering the torque to 3 to 6 pounds and emphasizing the importance of this tendon.

In acute situations, the integrity of the longitudinal arch may be initially maintained through static restraints. The valgus deformity of the hindfoot created by unopposed pull of the peroneus brevis through loss of the PTT secondarily causes loss of this integrity. According to Duchene⁴² this allows the gastrocnemius–soleus complex to act with a downward force at the talonavicular joint. Downward and medial pressure of the talar head stretches the calcaneonavicular ligaments. The plantar ligaments placed medially that unite the tarsus and metatarsus are comparatively much weaker than those on the lateral side. Eventually, the passive structures of the longitudinal arch give way under continued dynamic insult, and a flatfoot deformity results. In the beam theory, the natural curved beam of the bony architecture of the foot becomes straightened by repetitive tensile forces on its concave surface. In particular, the spring ligament is at risk of failure.

This view is supported by Niki and Sangeorzan¹⁴¹ who examined the progression to flatfoot deformity in a biomechanical study. The authors evaluated the sequential cause of the acquired flatfoot by creating a custom acrylic foot-loading frame to simulate heel strike, stance, and heel rise by altering tendon tension through regulated pneumatic cylinders. Simultaneously, axial compressive loads were applied to the tibias of these cadaveric specimens. Absence of posterior tibialis function was simulated simply by not activating the pneumatic cylinder attached to this tendon, while continuing normal cylinder load on all other tendons crossing the ankle. Thus, through cyclic evaluation, the authors studied the foot architecture with all tendons loaded in the absence of the posterior tibial tendon and, finally, with activation of the PTT (simulating repair). Small but statistically significant changes in the angular orientation of the bone architecture of the foot were noted after release of the posterior tibial tendon. These changes were not of the magnitude seen in a true flatfoot. This led the authors to surmise that the intact osteoligamentous structure of the foot is at least initially able to maintain normal alignment after acute PTT dysfunction. Of interest, when the PTT was restored in the flatfoot model, it did not restore the angular changes to anatomic magnitude. Again, these data support the importance of progressive longitudinal arch collapse through attenuation of the spring (and other plantar) ligaments.

The source of the attenuation may be directly related to a disorder within the gastrocnemius–soleus complex and its mechanical orientation. The valgus deformity of the hindfoot created by dysfunction of the PTT substantially alters the mechanical pull of the Achilles tendon. The Achilles tendon is placed lateral to the axis of the subtalar joint, allowing it to become an evertor of the hindfoot, accelerating the valgus deformity. Equally important, the moment arm of the plantar flexion force becomes the talonavicular joint rather than the metatarsal heads.⁵⁵ This proximal alteration in force concentration comes directly from the perpetual valgus hindfoot’s unlocking the transverse tarsal joint, eliminating the rigid lever of the foot at toe-off. This action accelerates attrition of the spring ligament with each gait cycle.

The intrinsic musculature attempts to compensate for the deficient arch by increased work. According to Mann and Inman,¹²¹ the intrinsics stabilize the transverse tarsal joint and thus create a more efficient lever. As expected, flatfooted persons require increased muscle action to maintain a rigid lever in light of the lack of support at other portions of the arch. Activity of the intrinsics thus begins at an earlier portion of the stance phase of gait, measured at 10% of the cycle rather than the normal 40%.

This cascade of events leads to a common pathway as the PTT fails and a flatfoot develops. Changes in the bone architecture are as those found by Niki,¹⁴¹ involving plantar flexion of the talus, eversion together with internal rotation of the calcaneus, and eversion of the navicular and the cuboid. Clinically, the hindfoot drifts into valgus and the forefoot into abduction. The posterior tibialis muscle fires earlier and longer to stabilize the hypermobile foot and to control the increased pronation and deviation. This increased demand on the muscle leads to fatigue. The deltoid ligament becomes involved late, as the severe flatfoot deformity places increased demand on the medial soft tissues, creating a valgus deformity and arthritic wear at the tibiotalar joint. Late-term effects occur as the calcaneus falls into further valgus. Abutment against the inferior tip of the fibula creates subfibular impingement. In addition, the relative shortening of the gastrocnemius–soleus complex leads to a permanent contraction, creating a relative equinus deformity of the ankle.

Etiology

Spontaneous rupture of the PTT was first suggested by Kettelkamp and Alexander⁹⁷ in 1969. Before that time, the authors were unable to locate a case of spontaneous rupture of the PTT in the literature. Controversy exists as to whether spontaneous rupture is truly spontaneous. McMaster¹²⁶ noted that spontaneous rupture did not occur in rabbits. In fact, even with a 75% iatrogenic laceration of the tendon, no ruptures occurred when normal stresses were applied. He stated that some form of disease process must be present to predispose the tendon to rupture.

Trauma of the magnitude to create a complete rupture of the PTT is rare. Funk⁵⁵ reported that of 19 patients treated for PTT tenosynovitis, only 4 could recall an inciting event. The authors agreed with McMaster,¹²⁶ stating that the lack of proximity of major trauma suggests that rupture of the PTT is more likely related to an intrinsic abnormality or biomechanical failure rather than an extrinsic traumatic factor.

Myerson noted in a group of elderly patients with rupture of the PTT that only 37 (14%) recalled a specific inciting traumatic event.¹³⁹ In contrast, Funk and Johnson⁵⁵ noted antecedent trauma in one half of the younger population in their study. Thus trauma to the tendon may be age stratified. This becomes evident from the case reports in the literature documenting acute rupture of the PTT resulting from ankle fractures. The first such report, by Giblin,⁵⁸ appeared in 1980: An interposed distal stump of a ruptured PTT prevented reduction of an isolated medial malleolus fracture. The patient had no prior symptoms of PTT dysfunction. In 1983, De Zwart³⁹ noted that two patients sustained rupture of the PTT at the level of a fractured medial malleolus in bimalleolar ankle fractures. A third report¹⁶⁹ suggested a common theme of a small flake of bone avulsed and visualized radiographically at the medial tibial metaphysis, just proximal to the medial malleolus fracture on oblique views.

A second theme¹⁶² suggests that acute rupture of the PTT is more commonly seen in ankle fractures caused by pronation and external rotation. The tight binding of the PTT by the flexor retinaculum contributes to acute rupture through sudden trauma. Fracture of the medial malleolus is not necessary,¹³¹ because the ruptured and interposed PTT may be seen with a deltoid ligament rupture associated with a pronation–external rotation fracture. This can even be seen in children,¹ raising the index of suspicion to prevent delayed flatfoot from developing after fracture of the ankle. Finally, there is some suggestion in the literature that acute injury to the PTT is the mechanism of disruption in dancers, rather than a slower degenerative process.³⁶ This article reviewed four dancers with either split tearing or more significant destruction of the PTT of acute onset. Often diagnosed initially with flexor hallucis longus tenosynovitis, the authors found this population did not present with a flatfoot, given their preexisting cavus consistent with high-level ballet dancers. Advanced diagnostic imaging was required to hone the diagnosis, and the pathology found at surgery suggested a more proximal location (directly posterior or superior to the medial malleolus) to the lesion than commonly seen in the more distal insidious ruptures.

More likely, repetitive microtrauma can lead to the indolent progression of symptoms through an inflammatory response that ultimately leads to tendon disruption. A tendon will not tolerate more than 1500 to 2000 cycles per hour. Tendon overloading can cause microtears that trigger an inflammatory response.⁷ Such microtears fail to heal, exacerbating the inflammatory response. As persons age, the tendon’s elastic compliance decreases through changes in the collagen structure, predisposing the tendon to damage.¹²⁴ Myerson’s¹³⁹ group of older patients reported the onset of symptoms at an average age of 60 years, seeking treatment at an average age of 64 years.

It is possible, however, that inflammation (preexisting tenosynovitis or tendinitis) has no involvement in disruption of the posterior tibial tendon. Mosier et al¹³³ evaluated gross and histologic specimens of 15 surgically resected posterior tibial tendons excised for rupture. Control tendons were those obtained from cadavers with no known disorder to the posterior tibial tendon, confirmed by gross inspection of the tendons at the time of harvest.

Direct inspection of the diseased tendons revealed a characteristic increased length from the malleolus to the insertion point when referenced with the cadaver tendons. Loss of normal tendon sheen and color was noted, and the tendon had a dull, white appearance. Incomplete longitudinal splitting was present without transverse rupture. Microscopic specimens stained appropriately revealed increased mucin content and myxoid degeneration. Excess mucin was found to alter the normal linear orientation of the tendon collagen bundles. Myxoid degeneration was consistent with a rupture within the substance of the tendon. At the insertion of the posterior tibial tendon, fibroblastic hypercellularity and chondroid metaplasia were present. Again, this had the effect of disrupting the linear collagen orientation. This haphazard or wavy configuration of the collagen leads to decreased tensile strength within the tendon, potentially leading to “spontaneous” rupture. This, of course, would validate McMaster’s concept of predisposition.¹²⁶ Most important, the authors found no signs of inflammatory infiltrates within either the tendon or the tenosynovium. This suggests that a degenerative condition, rather than an inflammatory one, creates disruption of the posterior tibial tendon.

An evaluation of the biochemical influences upon the PTT with reference to hormonal influence on PTT disease was explored by Bridegman et al.⁹⁰ The authors noted the higher propensity of PTT dysfunction in perimenopausal and postmenopausal women, leading to a concern that changes in hormone physiology might be the source of this observation. Evaluating both male and female patients through direct biopsy in those undergoing reconstruction, the authors used the adjacent flexor digitorum longus tendon as a control. Using a centrifuge to isolate tendon ribonucleic acid (RNA), followed by reverse transcription to isolate estrogen receptors in the specimens, the authors noted no difference in receptor expression between male and female patients. In addition, the diseased posterior tibial in female patients expressed a higher estrogen receptor level than the comparative control tendons. However, the sample size was not large enough to draw a statistically significant conclusion. A larger cohort will be required to discern this potential influence upon PTT dysfunction and the relative partiality towards female patients.

Trevino,¹⁸⁰ in 1981, performed the first histopathologic examination of diseased posterior tibial tendons. The authors found a stenosing tenosynovitis characterized by a loose, wavy configuration of the collagen. Other authors^38,135 confirmed this wavy configuration to the collagen, with irregular spaces between bundles. In fact, all of these investigators found little evidence of inflammation in patients with rupture of the posterior tibial tendon. Also, although no complete transverse ruptures were visualized, incompetence of the tendon in combination with elongation of the tendon was clear. Sutherland¹⁷⁵ suggested, through electromyographic studies, that the magnitude of elongation sufficient to reduce the ability of the PTT to act as a dynamic stabilizer of the longitudinal arch was as little as 1 cm. Clearly, degenerative changes manifested as intrasubstance collagen disruption is enough to create this length deficit.

Changes in the composition of the collagen matrix in disrupted posterior tibial tendons was explored by Goncalves-Neto et al.⁶² The authors confirmed the alteration in the normal linear orientation of collagen bundles in involved tendons. They noted neovascularization—an increase in size, number, and branching of blood vessels—in conjunction with an increased number of fibroblasts. These findings suggest attempts at repair. More important, they noted a shift in the type of collagen within the diseased tendon. Normal tendons are composed of type I collagen, with minor amounts of type III, IV, and V collagen. Incompetent posterior tibial tendons shifted the makeup of this collagen from 95% to approximately 56% type I. In addition, type III collagen increased 54%, and type V increased 26%. Investigators have found that as the percentages of type III and type V collagen increase within the extracellular matrix, the diameter of the collagen fibrils undergoes a marked reduction. Damaged tendon makes an attempt to heal itself, and it does so with this poorer quality collagen.

Even normal anatomic PTT exhibits alterations in collagen composition at specific locations. Petersen et al¹⁴³ found a change in composition of the superficial zone of the PTT directly adjacent to the pulley of the posterior medial malleolus. Using both light and transmission electron microscopy in combination with immunohistochemical methods, the authors noted increased type II collagen as well as acid glycosaminoglycans consistent with fibrocartilage rather than the standard composition of dense connective tissue. The authors believe the physiologic basis of this shift in composition is related to the character of the posterior tibial tendon’s changing from a traction tendon to a gliding tendon (subject to intermittent compressive and shear stresses) as it courses behind the medial malleolus.

The only portion of the PTT that exhibited composition consistent with fibrocartilage is the portion directly adjacent to the pulley, which is consistent with the most common site of spontaneous rupture. The fibrocartilage does not penetrate the entire diameter of the tendon at this location; rather, it remains superficial. This finding is physiologic, rather than pathologic, in the posterior tibial tendon. Thus it remains a point of weakness irrespective of the underlying disease process, but it remains vulnerable to any disease process affecting the tendon. From such an insult, a poor repair response occurs from the relatively avascular tissue. Repetitive stress or microtrauma can create such an insult because of the suspected lower resistance to tensile forces.

Despite this information, clinicians cannot eliminate the impact that inflammation has during the early stages of PTT dysfunction. Jahss first suggested a possible link between seronegative spondyloarthropathies and PTT tenosynovitis in 1982.⁸⁵ Seronegative disorders are inflammatory conditions generally occurring outside the synovium. They disproportionately involve sites of attachment of the capsule, ligament, and tendon to bone, known as entheses. These arthropathies are generalized and often involve multiple sites in the upper and lower extremities. Common concurrent connective tissue disorders include inflammatory bowel disease, psoriasis, urethritis, uveitis, conjunctivitis, and oral ulcers. Myerson¹³⁹ concluded that the majority of younger patients with PTT disorders have signs of systemic enthesopathies. This patient population was predominantly female. Symptoms of systemic disease occur at an average age of 27 years, and patients seek treatment at an average age of 39 years. Two thirds of the patients had, in addition to PTT tenosynovitis, other areas of inflammatory involvement. More than half had a first-degree relative with evidence of a connective tissue disorder.

Myerson¹³⁹ determined that two separate patient populations developed PTT tenosynovitis and subsequent dysfunction. Both populations underwent human leukocyte antigen (HLA) typing, supporting the association between age and seronegative spondyloarthropathies. Although only two patients in the older group had positive HLA markers, a majority of the younger population had HLA markers in their blood. Of particular interest is the Cw6 allele, the primary allele for psoriasis. Nearly half (47%) of the younger patient group had this allele compared with none in the older group. Only 12% in the institutional control group had Cw6 in their blood. This constellation of symptoms and laboratory results strongly suggests that inflammation plays a role in the inciting event of PTT dysfunction, at least in persons afflicted at a younger age. The younger population had a more rapid progression toward PTT rupture, encouraging prompt recognition of this potential cause and allowing earlier intervention. Stratifying the condition on an age-related basis has value only in patients with associated systemic conditions because degenerative tendinopathy has strong supportive evidence not only in Myerson’s older category of patients but also in the literature just noted.

In contrast to seronegative diseases that demonstrate an inflammatory cause inciting PTT dysfunction, rheumatoid arthritis has yet to show that it has a direct role in tendon destruction. Downey⁴¹ thinks that the chronic inflammatory mediators noted in rheumatoid arthritis are the predisposing factor for tenosynovitis. Michelson¹²⁸ suggested such a cause exists in as many as 64% of patients afflicted with rheumatoid disease. However, when he narrowed his criteria for diagnosis to include loss of a longitudinal arch, inability to perform a single-limb heel stance, and the inability to palpate the posterior tibial tendon, the incidence dropped to 11%.

The problem with all such studies is the concurrent destruction of the ligament complex at the subtalar and talonavicular joints known to occur in rheumatoid disease. This disorder confuses the specific cause of flatfoot. Any of the signs of PTT rupture noted by Michelson can be seen in patients with joint destruction from the synovitis of rheumatoid disease that creates flatfoot.

Jahss⁹⁴ examined the PTT in rheumatoid patients undergoing arthrodesis of the hindfoot for symptomatic flatfoot and found the tendon normal in appearance. Kirkham and Gibson¹⁰¹ studied 50 patients with rheumatoid arthritis and noted no instance of PTT dysfunction in those with progressive longitudinal arch collapse and hindfoot valgus. Finally, Keenan⁹⁵ performed electromyography on five patients with rheumatoid arthritis and hindfoot valgus and observed that muscle activity actually increased in this patient population versus controls. These data suggest that the PTT is actually overpulling in an attempt to correct the hindfoot valgus, rather than undergoing destruction by the underlying disease process. Thus attempts at linking the inflammation of rheumatoid arthritis to destruction of the PTT remain in question.

A vascular cause of PTT dysfunction has been suggested by numerous authors. Holmes and Mann⁸⁰ performed a review of 67 patients with a diagnosis of PTT rupture. This epidemiologic study found a statistical correlation between tendon rupture and both obesity (P = .005) and hypertension (P = .025). This patient population was older, ranging in age from 51 to 87 years, and the correlation remained strong in spite of the general population prevalence of these conditions in this age category. In addition, diabetes and steroid use (both oral and injection) were linked to PTT dysfunction.

All four associated conditions directly or indirectly compromise the blood supply to the posterior tibial tendon. Diabetes promotes vascular hyperplasia and sclerosis, leading to stiffness of the arterioles, luminal narrowing, and blood flow resistance. Steroids lead to local vascular attenuation, creating avascular tissue. Such ischemic insults can compromise an already tenuous blood supply to select portions of the posterior tibial tendon, leading to degenerative tendinopathy. This is supported by Kennedy and Willis,⁹⁶ who have shown through tendon-loading studies that steroid injections into the tendon sheath significantly weaken the tendon for up to 2 weeks. They found that the tendon disruption is related to collagen necrosis at the site of injection. Compromise in a tenuous blood supply can also occur in patients who have undergone previous surgery on the medial portion of the foot and ankle.

The PTT receives its blood supply primarily from the posterior tibial artery (although the insertion of the tendon might receive its blood supply from branches of the dorsalis pedis artery). There has been debate as to the significance of its aberrant blood supply contributing to tendon disorder. Frey and Shereff⁵⁴ used a modified Spaltholz technique to study the blood supply of the posterior tibial tendon. In doing so, they injected 28 cadaveric limbs with an India ink–gelatin suspension. This mixture was cleared via the Spaltholz technique, destroying the normal histology of the tendon while allowing visualization of the gross external and internal vasculature. To confirm isolated aberration in the PTT blood supply, they examined the adjacent flexor digitorum longus as a control.

Using this technique, Frey and Shereff⁵⁴ discovered that an important hypovascular zone (14 mm long) is present approximately 40 mm proximal to the insertion of the posterior tibial tendon.^80,174 This zone of relative avascularity generally begins at the medial malleolus as the tendon courses out of the groove. Of interest, no mesotenon was present at this level, and the visceral layer of the synovial sheath was hypovascular. In the control flexor digitorum longus tendon, no such zone existed, and consistent vascularity was noted throughout the tendon.

This hypovascular zone has a corollary in the supraspinatus tendon, where Rathbun and Macnab¹⁵² noted that tension in a select portion of the tendon can wring out the blood supply. Without an adequate blood supply, cells require surrounding extracellular fluid to provide nutrition through diffusion. The limits of diffusion are clear, however. Smith^165a suggests that nutrition will not be adequate when travel distance from source to destination exceeds 1 to 2 cm. Frey et al⁵⁴ thus reasoned that the hypovascular zone allows PTT deterioration with age.

There have been critics of this theory, however. In 2002, Petersen et al¹⁴⁴ used a different method to assess blood supply to the posterior tibial tendon. They criticize the Spaltholz technique as providing a large number of false-positive and false-negative results because of its subjective nature. Instead, they used a combination injection of ^99mTc, India ink, and gelatin, specifically studying the laminin. According to the authors, laminin is a basic component of the basement membrane, and staining with ^99mTc reliably and objectively detects blood vessels in the dense connective tissues.

Macroscopic assessment of the tendon at the level of the medial malleolus revealed an avascular, rather than hypovascular, segment. The majority of the avascular tissue was concentrated directly where the tendon is in contact with the malleolus. Confirmation was achieved through the immunohistochemical findings, which revealed an absence of laminin in this same location. Laminin was noted in the posterior quarter (farthest from the contact zone against the malleolus) of the PTT and proximal and distal to the malleolus.

Vascular theories explaining PTT degeneration and rupture have been challenged by Mosier et al.¹³³ The disorder of the collagen was discussed earlier. In addition to the increased mucin and myxoid degeneration, the authors noted neovascularization in the degenerative zone of the posterior tibial tendon. They surmised that the tendon cannot be avascular or hypovascular in these zones because the new blood vessels had to arise from existing circulation. In addition, they postulated that the fibroblast hypercellularity that they noted on the microscopic sections indicates increased metabolic activity within the presumed hypovascular zone, which is counterintuitive to the absence of vascular flow.

Two anatomic theories exist in the literature as to why the PTT is especially prone to developing long-standing inflammation and degeneration. The first potential agent is the overlying flexor retinaculum, which Jahss⁸⁵ suggests can cause compression and constriction of the tendon through synovial enlargement. This can impede circulation, fueling tendon degeneration. Second, the sharp turn or angle behind the malleolus creates excessive frictional forces with physical activity.¹⁷⁴ The excessive friction can contribute to the indolent inflammatory process.

In line with repetitive mechanical torque creating deficiency in the PTT are theories that congenital pes planus contributes to the disorder. Cozen²⁸ suggested that greater stress is placed on the PTT because of both the increased subtalar motion and medial column sag seen in patients with flexible flatfoot. This is supported by Dyal et al,⁴⁵ who reviewed radiographs of patients with PTT insufficiency and compared them with radiographs on the contralateral (asymptomatic) side. Using angular measurements defined by Bordelon¹⁰ and Sangeorzan et al,¹⁶⁰ the authors noted that 84% of asymptomatic feet and 86% of symptomatic feet had flatfoot deformities. These results were stratified to reveal that 23% of asymptomatic feet and 32% of symptomatic feet were moderately or severely flat. Interobserver reliability was highly correlated (P = .0001) among three surgeon observers, strengthening their results. Of course, this study does not prove that asymptomatic congenital pes planus is a potential cause of PTT insufficiency, but the suggestion is strong.

Indirectly, the contribution of a congenital flatfoot to PTT dysfunction is supported by Imhauser et al.⁸⁴ The authors created an in vitro model evaluating PTT function at heel rise. Unlike previous models, this construction allowed rotation of the tibia to contribute to a more physiologic gait pattern simulation. After creating a flatfoot deformity (through ligament sectioning), the authors noted a medial shift in load acting on the foot, placing undue stress on the PTT (and other medial structures) at heel lift. Such repetitive stress has already been discussed as a potential cause in PTT dysfunction.

This latter concept is supported by a 2006 paper by Uchiyama et.al.¹⁸¹ This study compared the gliding resistance of the PTT within its sheath both in simulated flatfoot and without pathology at different ankle positions. Gliding resistance mimics friction applied to the tendon in both static (standing) and dynamic (walking) situations. The authors extrapolated a validated technique developed to measure gliding resistance of tendons in the hand, which they found highly repeatable. Before flatfoot, the gliding resistance with the ankle in neutral averaged 77 N. After flatfoot deformity, this resistance increased to 104 N, on average. The authors postulated that this increased gliding resistance may accelerate tendon degeneration in both preexisting congenital flatfoot deformity and in progressive adult acquired flatfoot disorders.

Finally, the presence of an accessory navicular has a high correlation with developing PTT dysfunction.12,93

It is obvious that the cause of PTT insufficiency is most likely multifactorial and patient-specific. It is important for the clinician to assimilate the information discussed here to arrive at the correct underlying mechanism creating the disorder.